Jury Sentencing Survey
Kate Warner
Julia Davis
Maggie Walter
Rebecca Bradfield
Rachel Vermey
Report to the Criminology Research Council
Grant: CRC 04/06-07
April 2010
This is a project supported by a grant from the Criminology Research Council.
The views expressed are the responsibility of the author and are not necessarily those of the Council.
J
uries do not sentence offenders, but they are interested in the
outcome of cases they have tried, and they are well informed about
the circumstances of the particular case. The reaction of jurors to
sentences imposed on offenders is likely to influence public opinion.
It is also likely to provide a useful source of information to courts
about public opinion. If governments were concerned to know what
the public think about sentencing practice, a survey of the reactions
of jurors to sentences imposed in cases which those jurors tried
could provide interesting information. That could be a useful
practical test of whether there is some systematic failure of the
process to meet the expectations of the well-informed members of
the public.
THE HONOURABLE AM GLEESON AC (2005: 247)
ii
CONTENTS
List of Tables: ...................................................................................................................................... vi
List of Figures:................................................................................................................................................ viii
PART 1: AIMS, OBJECTIVES AND METHODOLOGY ................................................... 1
A. AIMS ............................................................................................................................................................. 1
B. APPROACH AND METHODOLOGY ................................................................................................... 2
1. Research design ................................................................................................................................. 2
Stage 1: Questionnaire 1 ................................................................................................................................... 3
Stage 2 ....................................................................................................................................................................... 4
Stage 3: Interviews .............................................................................................................................................. 4
2. Other sources of data ....................................................................................................................... 6
3. Approvals and timetable ................................................................................................................ 6
4. Analysis.................................................................................................................................................. 7
PART 2: LITERATURE REVIEW ....................................................................................... 8
A. WHY DOES PUBLIC OPINION MATTER?....................................................................................... 8
1. Explaining the rise of punitive penal policies and penal populism .............................. 9
2. The influence of the media ............................................................................................................ 9
3. Measuring and understanding public opinion, confidence, and punitiveness...... 11
4. Demographic factors ..................................................................................................................... 14
5. The impact of information on punitiveness ........................................................................ 14
6. Public views on the purposes of punishment ..................................................................... 17
7. Confidence in the Courts and the Judiciary ......................................................................... 18
8. Other questions measuring punitiveness............................................................................. 19
9. Social psychological literature on attitude formation and the concept of ‘public
opinion’.................................................................................................................................................... 20
10. Jury studies..................................................................................................................................... 21
Psychological literature on decision-making ....................................................................................... 22
11. Mechanisms for improving public knowledge about crime and sentencing ...... 22
12. The limits of public education using information .......................................................... 23
PART 3: RESULTS ............................................................................................................. 25
A. STAGE ONE ............................................................................................................................................ 25
1. Information on the cases in the sample ................................................................................ 25
Quantitative information ............................................................................................................................... 25
Appeals .................................................................................................................................................................. 26
Were sentencing submissions heard by respondents? .................................................................... 26
iii
2. Are jurors willing to be a source of public opinion on sentencing? .......................... 27
3. How representative are jurors of the general population?........................................... 30
4. From where do jurors source crime and sentencing information? ........................... 33
5. Crime victimisation ....................................................................................................................... 33
6. Juror’s proposed sentence compared with judge’s sentence ....................................... 34
7. Jurors’ knowledge of crime trends and sentencing patterns ....................................... 36
What is the relationship between crime and sentencing knowledge and sentencing
patterns and sentencing preference? ...................................................................................................... 40
8. Perceptions of risk of victimisation and fear of crime .................................................... 41
9. Jurors’ general opinion of current sentencing practices ................................................ 42
Do jurors’ opinions differ depending on the crime type of their trial? ..................................... 43
10. Jurors’ opinions of judges before sentence (Stage 1) ................................................... 45
B. STAGE 2 ................................................................................................................................................... 47
1. Who were the Stage 2 respondents? ...................................................................................... 47
Punitiveness of Stage 2 Respondents ...................................................................................................... 48
2. Jurors’ views of the judges’ sentences ................................................................................... 49
Judges sentencing remarks .......................................................................................................................... 54
3. Sentencing Goals ............................................................................................................................. 55
4. Aggravating and Mitigating Factors........................................................................................ 57
5. The Booklet ....................................................................................................................................... 59
6. Jurors’ general opinion of current sentencing practice after sentence.................... 61
Do jurors’ opinions differ depending on the crime type of their trial? ..................................... 63
Changing Levels of Punitiveness? .............................................................................................................. 64
7. Changes in jurors’ knowledge of crime and sentencing trends .................................. 64
8. Changes in jurors’ perceptions of the risk of victimisation and safety .................... 68
9. Jurors’ opinions of judges after sentence (Stage 2).......................................................... 68
10. Attitudes to public opinion, punishment and law breaking....................................... 71
11. Did jurors discuss the sentence or study with family or friends? ........................... 74
PART 4: DISCUSSION ....................................................................................................... 75
A. MAJOR FINDINGS ................................................................................................................................ 75
1. Jurors are willing to be used as a source of public opinion .......................................... 75
2. Jurors are reasonably representative of the general population ............................... 76
3. Informed public opinion is not as punitive as the populist view of public opinion
suggests ................................................................................................................................................... 76
4. Members of the public are more punitive when punitiveness is measured by
answers to abstract questions about sentencing than when asked about a sentence
in a particular case.............................................................................................................................. 78
Explaining the perception gap .................................................................................................................... 79
iv
Insights from the interviews ........................................................................................................................ 80
5. Informed members of the public do not consider judges are as ‘out of touch’ as
populist public opinion suggests .................................................................................................. 83
6. Jurors, in common with other members of the public, are poorly informed about
crime and sentencing......................................................................................................................... 86
7. The better informed (and the least fearful) are the least punitive ............................ 87
8. Information improved knowledge about crime and sentencing ................................ 87
9. In some respects punitiveness dropped after receiving more information but
respondents are not always consistently punitive................................................................ 88
10. Providing more information and improving knowledge is not a panacea........... 91
11. Jurors can act as conduits of information to the wider community ....................... 91
B. ANSWERS TO THE RESEARCH QUESTIONS............................................................................. 92
1. How can juries be utilised as a source of public opinion about sentencing? ......... 92
2. How receptive are jurors to learning about crime trends and sentencing? .......... 93
3. To what extent are jurors (as newly informed members of the public) satisfied
with the sentence imposed by the judge? ................................................................................. 93
4. What kind of information affects public satisfaction with sentencing? ................... 93
5. What variables affect jurors’ satisfaction with sentence? ............................................. 94
6. To what extent do the views of jurors as members of the public coincide or differ
from those of the judge as expressed in the sentencing comments? ............................. 94
C. CONCLUSIONS AND POLICY IMPLICATIONS ........................................................................... 95
1. Using jurors as a source of informed public opinion ....................................................... 95
2. Jurors can be used as conduits of information to better educate the general
public ........................................................................................................................................................ 95
3. The jury survey can be an effective strategy to counter public punitiveness ....... 95
References ....................................................................................................................................... 97
Appendix 1..................................................................................................................................... 104
Appendix 2..................................................................................................................................... 113
Appendix 3..................................................................................................................................... 129
v
LIST OF TABLES:
Table 1: Type of crime................................................................................................................. 25
Table 2: Most serious sentencing outcome imposed by judge (for all offenders) 25
Table 3: Juror response rate ..................................................................................................... 27
Table 4: Juror participation by trial judge........................................................................... 28
Table 5: Juror participation by place ..................................................................................... 28
Table 6: Juror participation by type of offence ................................................................. 29
Table 7: Juror participation by length of trial .................................................................... 29
Table 8: Juror participation by length of deliberation.................................................... 30
Table 9: Study jurors and Tasmanian jury eligible population ................................... 32
Table 10: Jurors’ perception of overall recorded crime trends .................................. 36
Table 11: Jurors’ perceptions of crime trends ................................................................... 37
Table 12: Perception of imprisonment rates for burglary and rape, percent ....... 39
Table 13: Perception of crime that involves violence and severity of sentence .. 41
Table 14: Crime victimisation rates for households and individuals ....................... 41
Table 15: Juror Perception of Risk of Victimisation ........................................................ 41
Table 16: Perceptions of safety ............................................................................................... 42
Table 17: Are current sentencing practices too tough/lenient?................................. 43
Table 18: General perceptions of sentencing leniency by respondent’s trial type
.................................................................................................................................................... 43
Table 19: Comparative severity of juror’s sentence by view on current sentencing
patterns ................................................................................................................................... 44
Table 20: Jurors view of in/out of touch and comparative sentence choice ......... 46
Table 21: Relationship between judicial remoteness and perceptions that
sentences are too lenient (Stage 1) .............................................................................. 47
Table 22: Participation by type of offence, Stage 1 and Stage 2 compared ............ 47
Table 23 Comparison on jurors who completed only Questionnaire 1 with those
that also completed Questionnaire 2 ........................................................................... 48
Table 24: Q2 - Juror’s response to judge’s sentence by sentence choice ................ 52
Table 25: Stage 1 comparative sentence severity compared with Stage 2 juror
response to judge’s sentence .......................................................................................... 54
Table 26: Ranking of sentencing goals ................................................................................. 55
Table 27: Aggravating Factors, relative importance, percent ..................................... 57
Table 28: Mitigating Factors, relative importance, percent ......................................... 58
Table 29: Jurors’ views on sentencing Stage 1 and Stage 2 compared .................... 62
vi
Table 30: Jurors’ views as to sentencing in general, Stage 1 and Stage 2 compared
(Stage 2 respondents only).............................................................................................. 62
Table 31: Jurors’ views as to sentencing in general, Questionnaire 1 responses
compared for Stage 1 only and Stage 2 respondents ............................................ 62
Table 32: General perceptions of sentencing leniency by respondent’s trial type
.................................................................................................................................................... 63
Table 33: Changes in jurors’ perceptions of overall recorded crime trends ......... 65
Table 34: Reading the booklet and assessment of overall crime trends ................. 65
Table 35: Changes in jurors’ perceptions of crime trends for selected offences . 65
Table 36: Changes in perception of proportion of crime that involves violence . 66
Table 37: Reading the booklet and correct assessment of the proportion of crime
that involves violence ........................................................................................................ 66
Table 38: Changes in perception of imprisonment rates .............................................. 66
Table 39: Use of the booklet and knowledge of burglary imprisonment rate ...... 67
Table 40: Knowledge of imprisonment rates and education level of respondent67
Table 41: Estimation of the risk of victimisation for various offences, percent. .. 68
Table 42: Perceptions of safety ............................................................................................... 68
Table 43: The Relationship between perceptions of judicial remoteness and the
appropriateness of the sentence. .................................................................................. 69
Table 44: Relationship between perceptions of judicial remoteness and general
attitudes to sentences (Stage 2) .................................................................................... 70
Table 45: The Relationship between perceptions of judicial remoteness and
sentence choice at Stage 1................................................................................................ 71
Table 46: Relationship between perceptions of judicial remoteness and sentence
preference at Stage 2.......................................................................................................... 71
Table 47: Attitudes to public opinion, punishment and law breaking (percent). 72
Table 48: Agreement with death penalty for murder by agreement that judges
should reflect public opinion in sentencing, percent ............................................ 72
Table 49: Agreement that judges should reflect public opinion in sentencing by
choice of more or less severe sentence at Stage 1 .................................................. 72
Table 50: Agreement that judges should reflect public opinion in sentencing by
sentence preference at Stage 2. ..................................................................................... 73
Table 51: Agreement that judges should reflect public opinion in sentencing by
how in touch judges are with public opinion, percent.......................................... 74
Table 52: Whether discussed the sentence by sentence preference at Stage 2 ... 74
vii
LIST OF FIGURES:
Figure 1: Sources of information about crime and sentencing ................................... 33
Figure 2: Type of crime committed against victims ........................................................ 34
Figure 3: Judge and juror’s sentence compared by type of Offence .......................... 35
Figure 4: Rate for recorded crime, Tasmania, 1981-82 to 2005-06. ........................ 37
Figure 5: Recorded crime in Tasmania: distribution of offences and perceptions
of crimes of violence ......................................................................................................... 38
Figure 6: Perceptions of proportion of crime that is violent ....................................... 38
Figure 7: Sentences types by offence 2001-2006............................................................. 40
Figure 8: How in touch are judges with public opinion in relation to sentencing46
Figure 9: Severity of juror’s proposed sentence compared with judge’s sentence
(percent) by type of offence ............................................................................................ 49
Figure 10: Was the sentence expected by jurors? (by offence type) ........................ 50
Figure 11: How appropriate was the sentence for each crime type ......................... 50
Figure 12: Satisfaction levels and sentence choice (all offences) .............................. 51
Figure 13: Stage 2 comparison sentence severity by offence type ............................ 53
Figure 14: Most important sentencing goal for sex offenders .................................... 55
Figure 15: Most important sentencing goal for violent offenders ............................. 56
Figure 16: Most important sentencing goal for drug offenders .................................. 56
Figure 17: Most important sentencing goal for property offenders ......................... 56
Figure 18: How well was the booklet read? ....................................................................... 59
Figure 19: Juror plans for booklet .......................................................................................... 60
Figure 20: Mean ratings for the booklet on various dimensions ............................... 60
Figure 21: Impact of sentencing comments ....................................................................... 61
Figure 22: How in touch are judges, changes in response (all respondents) ........ 69
viii
PART 1
AIMS, OBJECTIVES AND METHODOLOGY
A. AIMS
The study has three immediate aims:

To investigate a new method of ascertaining public opinion by assessing the
feasibility of using juries as a source of informed public opinion.

To develop a new way of improving public knowledge about sentencing by
using jurors as conduits of information.

To ascertain attitudes to sentencing from an informed sector of the public.
The broader aim of the study is to counter populist penal punitiveness by addressing
the „comedy of errors‟; namely, the situation that criminal justice policy and practice
is not based on a proper understanding of public opinion, and public opinion is not
based on a proper understanding of policy and practice. Public opinion surveys
conducted across the world over the last four decades consistently find that between
70 and 80 percent of respondents think that sentences are too lenient. More
sophisticated research has led researchers to label this „a methodological artefact – a
result of the way in which public opinion has been measured‟ (Gelb 2008a: 45). It has
been found that people have little accurate knowledge of crime and the criminal
justice system, that those who have better knowledge are less punitive and that when
given more information, people become less punitive. This suggests that a strategy to
counter penal punitiveness is to improve public knowledge about crime and
sentencing matters and to devise better methods of ascertaining informed public
opinion. The provision of a better measure of informed public attitudes (in contrast to
uninformed and flawed public opinion polls) will provide the basis for a reasoned
argument for politicians and policy advisers to use when resisting calls made by the
popular print and broadcasting media to increase penalties and to get tough on crime.
Providing a source of informed public opinion, which can be fed into the criminal
justice system, has the potential to improve public confidence in the system. Because
of the relationship between ratings of confidence in the courts and perceptions of
severity – those who report that sentences are too lenient are less confident in the
courts – improving confidence in the courts can also reduce punitiveness.
The Research Questions: The following six research questions were formulated:
1.
How can juries be utilised as a source of public opinion about sentencing?

Do they have the willingness and capacity to participate in a study
exploring their views on sentencing?

How willing are jurors to respond to invitations to stay and listen to
sentencing proceedings?

Do they have the willingness to read and the capacity to understand
briefing information about sentencing?
1
Jury Sentencing Survey

Are jurors willing to complete a survey form about sentencing?

Are jurors willing to respond to requests to be interviewed about their
views?
2.
How receptive are jurors to learning about crime trends and sentencing?
3.
To what extent are jurors (as newly informed members of the public)
satisfied with the sentence imposed by the judge?
4.
What kind of information affects public satisfaction with sentencing?
5.
6.

Listening to the sentencing submissions?

Knowledge of crime trends?

Information about sentencing law and sentencing patterns?
What variables affect jurors‟ satisfaction with sentence?

Variables relating to juror demographics?

Variables relating to the offence type?

Variables relating to the offender?

Variables relating to the victim?
To what extent do the views of jurors as members of the public coincide or
differ from those of the judge as expressed in the sentencing comments?
B. APPROACH AND METHODOLOGY
1. Research design
The research design envisaged that all jurors in trials with a guilty verdict would be
surveyed over a period of two years. It was estimated that two years would produce
some 150-160 trials with guilty verdicts. Recruitment of jurors began in midSeptember 2007, and finished in early September two years later. In this period there
were 162 trials returning guilty verdicts. Approval for the study was sought and
obtained from the Chief Justice and the three-stage research design was developed in
consultation with him. The Attorney-General‟s approval was also sought. Although
jurors were not asked to divulge the content of their deliberations at any stage, it was
decided to ask for exemption from disclosing prohibited matters under the Juries Act
2003 s 58(6)(e) in the event that a juror did disclose something in relation to their
deliberation in the course of an interview. The Attorney-General granted approval for
the research. Before the project began, the project team leader briefed judges about
the project at a Judges‟ Meeting. Relevant court staff were also briefed by a circular
prepared by the Chief Justice and the Research Team.
The project tracked the attitudes of participants at three stages:
Stage 1 – Initial response: after verdict and before sentence via responses to a survey
attached to the study consent form.
2
Part 1: Aims, Objectives and Methodology
Stage 2 – Informed response: after reading the briefing materials and sentencing
remarks via a second survey form.
Stage 3 – Considered response: after completing the informed-response survey form
via a personal interview with selected numbers of participants (qualitative analysis).
Stage 1: Questionnaire 1
Each time a new jury panel was summoned to the court in Hobart, Burnie and
Launceston, the sitting senior judge briefly introduced a member of the research team
to the panel expressing the Court‟s approval and support for the project. The member
of the research team then outlined the aims of the study and explained that in the
event of a guilty verdict, jurors would be invited to participate by staying to hear the
sentencing submissions and then complete a Questionnaire, which was described as
Stage 1 of the Study. They were also invited to take part in Stage 2 and to agree to a
Stage 3 interview. What each stage involved was briefly explained. There were up to
ten jury panels each year in each city. In Hobart, the research team member(s) was
always introduced by the judge. However, this did not always happen in Burnie and
Launceston where the introductions were sometimes done by the jury officer if the
judge was not able to do so. For the last nine months of the study there was no
research team member available in Burnie to explain the project, so this was done by
the jury officer.
After a guilty verdict, the trial judge invited the jury to participate in the study by
staying behind to listen to the sentencing submissions. At the conclusion of the
submissions the judge then asked the jury to leave the court with the jury officer to fill
in Questionnaire 1. In most cases the handing down of the sentence was adjourned to
a later date. However, if it was to be handed down immediately, the jurors left the
court before this took place. While the initial plan was for the jurors to complete the
Questionnaire before they left the court buildings, in many cases, particularly in the
evenings, the court staff preferred to give the jurors the option of completing the
questionnaire immediately or taking it with them and posting it to the University in a
pre-paid envelope. In an estimated 10 percent of cases the sentencing submissions are
adjourned. In this case jurors completed the questionnaire without hearing the
sentencing submissions, which were instead transcribed and sent out with
Questionnaire 2.
Consent forms were distributed with Questionnaire 1. Questionnaire 1 asked a series
of questions covering the following:

An indication of what the sentence should be (for example, a sentence of
imprisonment of X length or a specified non-custodial penalty or don‟t
know).

Some questions about crime rates and perceptions of risk of victimisation.

An indication of whether current sentencing practices are generally too
lenient, harsh etc, and whether judges are in touch with public opinion.

Demographic details.

Contact details (postal address and email or phone number) so that „briefing
materials‟ including the sentencing remarks, a booklet and a second
questionnaire could be sent to the juror.
3
Jury Sentencing Survey
A copy of Questionnaire 1, the consent form, information sheet and an additional
page that was inserted into Questionnaire 1 where there were multiple offenders, is
included in Appendix 1. Jury officers were briefed as to what they should say to the
jury. They were instructed to tell the jury not to worry about what their first estimate
of the sentence should be and to emphasise that they would have the opportunity to
revise their view in the second Questionnaire after they had been sent information
about crime and sentencing, including the sentencing comments given by the judge.
High participation rates depended on the jury officer or court staff having the
Questionnaires, consent form and information sheet ready for distribution and
handing them out to those jurors who indicated a readiness to participate.
Stage 2
The standard package, which was sent to all jurors who agreed to participate in Stage
2, included:

A booklet containing some general information on crime rates and
sentencing matters (See Appendix 2).

An insert with data on sentencing patterns about the particular crime for
which the defendant had been convicted (Inserts, see samples in Appendix
2).

The sentencing comments.

A transcript of the sentencing submissions if these were adjourned in whole
or part (these were supplied by the Court).

Questionnaire 2 and a pre-paid and addressed envelope. There were two
versions of Questionnaire 2: one for cases with one defendant only and one
for multiple defendants. Copies of both versions of Questionnaire 2 are
included (See Appendix 3).
Stage 2 participants were asked to complete Questionnaire 2 and return it to the
research team after they had read the briefing materials package. Questionnaire 2
included four sections:
Section A: questions about the sentence in the case, the sentencing remarks,
sentencing goals and aggravating and mitigating factors.
Section B: questions about the information package.
Section C: Views about sentencing in general, crime trends and perceptions of
risk of victimisation.
Section D: discussion of the sentence or booklet with others.
At the end of the questionnaire respondents were asked to specify whether they were
willing to be interviewed. If no response was received in about 21 days, jurors were
sent a duplicate package.
Stage 3: Interviews
From the group of jurors willing to participate in interviews, it was planned to
select 50 for semi-structured in-depth interviews. The intention was to pick a
spread of jurors from Hobart, Launceston and Burnie and to include at least 15
whose opinions had changed and become harsher; at least 15 whose opinions had
4
Part 1: Aims, Objectives and Methodology
changed and become more lenient; and at least 15 whose opinions had remained
the same. These were to be selected from a range of offence types. As the study
progressed, however, it became apparent that we were not going to be able to find
15 participants whose response to the general question about sentence severity had
become unambiguously more punitive. This was because most participants
appeared to become more moderate at Stage 2. Consequently it was decided to
select jurors so that there was a spread of both offence types and jurors in terms of
their sex, age and background. Furthermore, several jurors who had heard more
than one case were selected so that the interviews could explore any differences in
those jurors‟ attitudes to different kinds of offenders and offences. In total, 50
jurors who had participated in 62 trials were interviewed:
23 trials involved crimes of violence;
20 trials involved sex offences;
7
trials involved property offences;
10 trials involved drug offences; and
2
trials were classified as „other‟.
The interviews loosely followed the structure outlined below:
The sentence selected: The interviews generally began with a discussion of the
sentence and an exploration of the juror‟s view of how appropriate the judge‟s
sentence was. Jurors were asked why they favoured their chosen sentence, the reasons
for any change in view, the goals they thought the selected sentence would achieve
and for their response to the aggravating and mitigating factors. If a suspended
sentence was imposed in the case, they were asked their views about this sentencing
option. The interview also explored their views of the offender in the case. We
attempted to get jurors to give their particular case a seriousness score out of ten.
The media response: Jurors were asked if they were aware of whether the media had
reported on their trial, and if so, had seen the coverage, and did they believe the media
had accurately portrayed the story.
Views about sentencing in general and judges: In this part of the interview, we
explored general views about sentencing levels in the abstract and any changes that
may have occurred in the juror‟s view between Questionnaire 1 and 2. We
endeavoured to explore the kinds of cases they were thinking of if they said sentences
for violent offences were too lenient. Where a juror agreed with the judge‟s specific
sentence in the case of an offence of violence, but had also stated that in general they
thought sentences for violent crimes were too lenient, we explored the reasons for the
dichotomy in views. We asked jurors for their views about the desirability of
structuring the sentencing decision-making process by reducing judicial discretion
and introducing grid systems. They were also asked whether they thought jurors
might like to be more involved in the sentencing process.
We investigated the jurors‟ responses to the question: „how in touch do you think
judges are with public opinion on sentencing‟, by exploring any change between
Questionnaire 1 (B6) and Questionnaire 2 (C3) on this issue and asking them to
elaborate upon what they meant by „somewhat in touch‟ for example. In some cases
we linked their response to this question with the question on whether judges should
reflect public opinion when sentencing (Questionnaire 2 C10). We tried to explore the
impact of the jury experience on their confidence in the criminal justice system.
5
Jury Sentencing Survey
Juror response to the study: To explore the reasons why the response rate differed
between different trials, we asked jurors about the general juror interest in the case,
the jury dynamics, length of the trial and their deliberations, and any possible feelings
of intimidation. This was particularly valuable when jurors had sat on more than one
case and had participated in the study by filling out the questionnaires about one case
but not the other.
The booklet: We asked about the booklet and its value; whether they thought jurors
and the public would be interested in accessing sentencing remarks. Finally jurors
were asked if they had any questions or comments on the jury experience.
2. Other sources of data
In addition to the data from Questionnaire 1 and 2, we gathered information from the
courts on all trials with guilty verdicts to ensure that some data from trials with a nil
response rate was also collected. To assist in the analysis of the response rate, we also
obtained data on the length of each trial and the time of verdict.
Local newspapers were searched for coverage of the trials so that any relevant
newspaper story could be discussed with jurors interviewed.
To provide the sentencing data for the Crime and Sentencing booklet and the data
sheets for the information package sent out at Stage 2, sentencing data was compiled
from the Supreme Court database of sentencing comments into an Excel database of
Supreme Court sentences for the years 2001-2006. The Supreme Court database does
not have the capacity to show cumulative data on sentencing outcomes or the
sentencing ranges for particular crimes. As the study progressed, data from 2007 and
2008 was added to the database and the data sheet inserts were prepared when needed
to match the data required for a particular case. Approximately 80 data sheets were
prepared for this purpose.
3. Approvals and timetable
It was initially proposed to start recruiting jurors in June 2007. However, delays with
the finalisation and signing of the deed delayed the start. Questionnaires were drafted,
and the consent forms and the information sheet prepared. Ethics approval was
granted by the Chair of the University‟s Human Research Ethics Committee
(Tasmania) on 20 June 2007 (Jury Sentencing Survey No 04/06-07 reference H9487).
Approval to proceed was also obtained from the Attorney-General. The Juries Act
2003 (Tas) s 58 prohibits soliciting or obtaining the disclosure by a juror or former
juror of statements made and opinions expressed in the course of deliberations. None
of the questions we proposed to ask related to jury deliberations. Rather, they focused
on the individual juror‟s views on matters relevant to sentence, such as the matters
they considered aggravated and mitigated offender culpability and offence
seriousness. Nevertheless, approval was sought from the Attorney-General for the
project under s 58(6)(e). In the interview stage, jurors sometimes volunteered
information about the deliberations and as the Attorney-General had approved the
project, it was decided jurors could be asked if they had discussed the possible
sentence in the course of their deliberations.
The questionnaires were piloted in August 2007 and amendments were made to some
of the questions as a result. In particular, the pilot participants resisted answering a
6
Part 1: Aims, Objectives and Methodology
general question in relation to whether current sentences are too tough, about right or
too lenient. As a result this question was omitted and it was asked in relation to crime
types separately (See Q1 B4). Feedback on the booklet also resulted in some
simplification and clarification.
A graphic designer was employed to style the Questionnaires and the Booklet before
the final printing.
Recruitment of jurors began on 17 September 2007 and the last jury panel to be
included in the study was the panel which was addressed on 18 August 2009. The last
trial with a guilty verdict in the study concluded on 1 September 2009. Final followup Questionnaires were sent out on 16 October 2009 and the cut-off for return of
Questionnaire 2 was 1 December 2009.
Interviews began in late 2007 and by September 2009, 50 interviews had been
conducted by three project members. Interviews lasted between 40-90 minutes and
were conducted either by single interviewers or by pairs of interviewers. All jurors
consented to the recording of the interviews, which were recorded and later
transcribed.
4. Analysis
A Statistical Package for the Social Sciences (SPSS) database was established for the
data from Questionnaire 1 and 2 in December 2007, and data was entered as it was
received. Preliminary analysis was done on the data in December 2007, in August
2008 and in July 2009. The first analysis was done to present some preliminary results
at two conferences in early 2008 and the second analysis was completed for an AIC
Trends and Issues Paper. As a result the database was refined and additional fields
added.
Once the interviews had been transcribed and analysed, the jurors were allocated code
names using the Nato Military Phonetic Alphabet (Alfa, Bravo, Charlie, Delta etc) to
preserve their anonymity. In this report the interviews are used to complement the
discussion of the quantitative findings in Part 4.
7
PART 2
LITERATURE REVIEW
The literature that is relevant to this study is broad and covers a number of disciplines
including law, criminology, sociology and psychology. This review is a brief survey
of this literature.
A. WHY DOES PUBLIC OPINION MATTER?
Central to the research questions explored in this study is the issue of gauging public
opinion. A preliminary question is why does public opinion matter? The answer to
this question is threefold:
First, it matters because it is linked with public confidence in the criminal justice
system and critical public institutions (Indermaur and Roberts 2009). Sentencing is
the most visible aspect of the criminal justice system and public attitudes to it have a
considerable impact on the state of public confidence in the criminal justice system.
When asked why they have little confidence in the courts, people typically cite lenient
sentencing (Gelb 2008a: 3; Roberts, Crutcher and Verburgge 2007: 84). In turn,
public confidence in the criminal justice system matters because it affects the
functioning of the system itself: a lack of confidence in the system can reduce the
reporting of crime, and may inhibit the co-operation of witnesses and attendance of
the public for jury service (Roberts 2007). Moreover, Canadian research shows that
there is a correlation between public institutions and a sense of belonging, suggesting
that confidence in critical public institutions, such as the criminal justice system,
promotes social cohesion (Roberts 2007: 155).
Secondly, it is widely recognised that sentencers (judges and magistrates), as well as
policy makers, should have regard to informed public opinion. For example, the
seriousness with which society regards a particular offence is something sentences
should reflect (Mason 2002; Mackenzie 2005: 138-148). This issue is one which has
been considered extra-judicially by a number of judges including Sir Anthony Mason
(2002). For example, in 2004, Chief Justice Murray Gleeson (Gleeson 2005) as he
was then, acknowledged that judges are expected to know and be responsive to public
opinion. He then posed a series of questions: How should they keep in touch? Should
they employ experts to undertake regular surveys of public opinion? What level of
knowledge and understanding of a problem qualifies people to have opinions that
ought to influence judicial decision-making? Who exactly is it that judges should be
in touch with?
Thirdly, the public has become a key factor in shaping penal policy (Roberts and
Hough 2005: 160; Ryan 2005: 145). Direct political pressure on decision makers to
accommodate public opinion is increasing. Ordinary people, it seems, want more
ownership of their democracy than in the past (Ryan 2005: 145). The four pillars or
axes of justice are the state, the offender, the victim and now the public (Freiberg
2003). Public opinion can have an impact on criminal justice policy development,
8
Part 2: Literature Review
forcing changes to the laws (Roberts 2007). In other words „sentencing matters‟
(Tonry 1996).
1. Explaining the rise of punitive penal policies and penal populism
This section of the review expands on the third point above. Heightened sensitivity to
public opinion underlies „penal populism‟ (or „populist punitiveness‟) a term coined
by Bottoms to describe the „notion of politicians tapping into and using for their own
purposes, what they believe to be the public‟s punitive stance (Bottoms 1995: 40). It
does not simply refer to political responsiveness to popular views, but embodies the
idea of taking public expressions of punitiveness at face value and advancing policy
without regard to its effects. Misinformed public opinion is exploited to win votes.
„Penal populists allow the electoral advantage of a policy to take precedence over its
penal effectiveness.‟ (Roberts et al. 2003: 5).
One stream of theorising has explained the rise in punitive policies in the concerns of
late modernity. A convergence of social, cultural, economic, technological, and
ecological change has created increasing public scepticism about the ability of
national states to regulate change through the political process. The general insecurity
people feel in the face of such change translates to concerns about crime and personal
safety. This explanation for punitiveness revolves around the broad social anxieties
besetting the middle class (Roberts et al. 2003: 68-75). An environment is created in
which the „criminology of the other‟ can flourish (Roberts et al. 2003: 68, citing
Garland 2000). David Garland (2001) and others such as Aas (2005), have argued that
the influence of the expert in criminal justice policy has declined and been replaced
by the voice of the public. Research and criminological knowledge has been
downgraded and in its place is a new deference to the authority of „the people‟,
common sense and „getting back to basics‟ (Garland 2001: 20). „From the perspective
of populist discourse, criminological discourse is discarded as elitist, as high
knowledge, distant from people‟s feelings‟ (Aas 2005: 151). This is a theme pursued
by Pratt (2002) who claims that a new axis of penal power has emerged „in which the
indifference of the general public is increasingly giving way to intolerance and
demands for still greater manifestations of repressive punishment‟ as well as more
„ostentatious and emotive‟ forms of punishment (Pratt 2002: 182, cited in Gelb 2006:
5). The democratisation of punishment has its downsides.
2. The influence of the media
The role of the mass media in influencing public opinion and punitive policies is a
theme in Garland‟s work and that of many others. As Garland (2000: 363) succinctly
stated, „[i]t has surrounded us with images of crime, pursuit and punishment, and has
provided us with regular, everyday occasions in which to play out the emotions of
fear, anger, resentment, and fascination that crime provokes.‟
Indermaur and Roberts found that only five percent of the Australian Survey of Social
Attitudes (AuSSA) survey respondents in 2007 reported any contact with a criminal
court in the previous year (2009: 3). Most people do not have direct access to firsthand information about the criminal justice system, either through personal
experience or even the experience of family and friends. Instead, they tend to learn
about it through the mass media outlets, which play an integral role in the
construction of public opinion and the public reality of crime (Gelb 2006: 15). In
9
Jury Sentencing Survey
reporting crime, the media is quick to seize upon lenient punishment of offenders and
use it as a basis of criticism of judges. Too often politicians capitalise on these
criticisms when a law and order campaign offers the prospect of political advantage
(Mason 2002). Judicial criticism of media reporting of crime has a long history. Over
100 years ago, Stephen noted that „[n]ewspaper reports are necessarily much
condensed, and they generally omit many points which weigh with the judge in
determining what sentence to pass‟ (Stephen 1883: 90, cited in Roberts and Stalans
1997: 216; Gelb 2006: 15). More recent Australian judicial critics and observers of
media crime reporting include Gleeson (2007), Spigelman (2005), Sackville (2005,
2009) and Lasry (2009).
A great deal of research has explored the way that crime is represented in the news
media (Roberts et al. 2003: 77). For example, a study by Graber (1980) found that 25
percent of all crime stories were on murder, although this crime constitutes less than
one percent of recorded crime (cited in Roberts and Doob 1990; Gelb 2006: 15). Only
a small proportion of sentences involve imprisonment but the Canadian Sentencing
Commission found that 70 percent of media reports focus on this sentencing outcome
(Gelb 2006: 15).
The way the media creates conditions for a conservative and punitive response to
crime has been discussed by many writers (Kennamer 1992; Lovegrove 1998;
Indermaur 2000; Bloustein and Israel 2006; Casey and Mohr 2005; Indermaur and
Roberts 2005: 148; Schulz 2008; and the authors reviewed in Roberts et al. 2003: 7692). Gelb explains how the media influences the public to perceive sentencing as too
lenient:
As people are overly influenced by single-case information, people falsely
generalise that leniency characterises the entire sentencing process. The media
tend to focus particularly on violent crime, which provides a disproportionate
emphasis on this type of crime relative to its prevalence in the community.
People then perceive this type of event as typical, which affects both their
knowledge of the facts about crime as well as their general levels of fear of
crime. Both of these in turn have been shown to influence perceptions of
leniency in sentencing (Gelb 2006: 15).
A recent Australian study of 300 media headlines found that a distinct pattern of
disrespect and disapproval of judicial sentencing was connected with a „discourse of
direction‟ and demands for increases in penalties (Schulz 2008). Beale (2006) offers
two explanations of how the media are able to cause the public to perceive crime to be
more serious than it actually is. First, by agenda setting, which involves directing the
public‟s attention to certain issues. The second is by priming, which describes the
media‟s ability to affect the criteria by which viewers judge public policies and public
officials.
Despite the fact that the source of most people‟s information about crime and the
criminal justice is the media, most also acknowledge that the media do not provide
accurate information (Square Holes 2006 cited in Gelb 2008a: 6).
Survey research has investigated the ways that people gain information about
sentencing. A 1986 Canadian Sentencing Commission survey found that 95 percent of
people derive their information from the news media (cited by Gelb 2006: 15). In
Indermaur‟s 1990 Perth study, the main source of information about court practices
was the media (cited in Gelb 2006: 31). Survey research has also correlated the types
of newspaper readers with punitiveness and has found that tabloid newspaper readers
10
Part 2: Literature Review
tend to be more punitive (Hough and Roberts 2007). Respondents in the AuSSA 2007
survey identified the media as their most important sources of information, with
almost eight out of ten rating TV, radio and newspapers as „fairly or very important‟
(Roberts and Indermaur 2009: 9).
The fact that most members of the public derive their information about crime and
sentencing from the media has been suggested as a reason why public opinion surveys
find that the justice system typically attracts poorer ratings than does the health care
system, the educational system or the military (Hough and Roberts 2004). Julian
Roberts explains that, while almost everyone has contact with the health or education
system, few have direct experience with the criminal courts. Information about the
courts is therefore filtered through the news media, which focus on the negative
aspects of criminal justice, such as failed prosecutions or lenient sentences (Roberts
2007: 165).
3. Measuring and understanding public opinion, confidence, and
punitiveness
Public opinion can be measured by media, polls, representative surveys, focus groups
and deliberative polls, and each method has its own advantages and disadvantages
(Gelb 2006). It seems a comprehensive picture of public opinion/public judgment can
only be obtained by a multi-method approach (Roberts and Stalans 1997). The
Victorian Sentencing Advisory Council advocates the development of a flexible „suite
of methodologies‟ that can be used to answer different kinds of research questions
(Gelb 2006: 41). There is literature on such methodological issues as the effects of
question order on responses and the design of optimal survey questions (Gelb 2008b)
as well as literature on focus groups and deliberative polls (Luskin, Fishkin and
Jowell 2002; Connelly, Wagner and Jones 2002; Yankelovich 1991).
Media and representative polls show that, in the abstract, the public thinks that
sentences are too lenient. Over three decades and across several countries (from North
America and Australia to the United Kingdom) about 70-80 percent of respondents
reported that sentences are too lenient with slightly lower rates for Canada in recent
years (Gelb 2006, 11). For example, in Indermaur‟s 1987 Perth study, 76 percent of
the sample thought sentences were not severe enough, 19 percent thought they were
about right and five percent said they were too severe (Indermaur 1987). The
exception to this pattern appears to be Singapore, where only five percent of the
public polled held the view that sentences were too lenient and three quarters
expressed the view that they were just right (Roberts 2007: note 15). Although
increases in the imprisonment rate and the implementation of numerous examples of
punitive policies suggest that sentencing policy has in many respects become more
punitive over the last two or so decades, public punitiveness itself (as measured by the
response to the statement that people who break the law should be given stiffer
sentences) has been quite stable or declining (Indermaur and Roberts 2005: 155).
The limitation of using a single question in surveys has been recognised and
addressed by adding a follow-up question that asks respondents about the kind of
offender that they had in mind when they gave their first response. This has shown
that most people were thinking of a violent or repeat offender and those who think
sentences are too lenient were most likely to be thinking of violent offenders (Doob
and Roberts 1983, cited in Gelb 2006: 12-13; Indermaur 1987, cited in Gelb 2006:
11
Jury Sentencing Survey
30). Roberts and Stalans (1997) recommend that the question be asked twice – once
for non-violent offenders and once for violent offenders. In their Canadian survey this
revealed that 80 percent of respondents thought sentences for violent offenders were
too lenient but less than half thought so for non-violent offenders (Roberts and Stalans
1997: 208, cited in Gelb 2006: 13).
An important finding from representative surveys is that people have very little
accurate knowledge about crime and the criminal justice system. For example, people
tend to:

perceive crime to be constantly increasing;

over-estimate the proportion of recorded crime that involves violence; and

under-estimate the severity of sentencing practices for specific offences
(Gelb 2006: 13, 30).
The AuSSA 2007 survey confirms that in Australia:
A large majority of the public have inaccurate views about the occurrence of crime
and the severity of sentencing. Consistent with previous Australian and international
research, the Australian public perceives crime to be increasing when it isn‟t,
overestimates the proportion of crime that involves violence and underestimates the
proportion of charged persons who go on to be convicted and imprisoned (Roberts
and Indermaur 2009: ix and see also 24).
The British Crime Survey (BCS) has, since 1996, asked respondents about levels of
crime nationally and locally (Thorpe and Hall 2009: 96). Since 2004-05 the gap
between perception of change in national and local crime levels have widened. So
while in 2008-09 there had been a continued decrease in the proportion of people who
thought crime has increased locally (to 36 percent) there had been a marked increase
in the proportion of people who thought that crime had increased nationally (from 65
percent in 2007-08 to 75 percent in 2008-09). To better understand the divergence
between national and local perceptions of crime trends, the 2008-09 BCS included
questions about specific crimes. It found that the proportion of people perceiving an
increase in crime nationally was higher for crimes that attract most media coverage,
such as knife and gun crime, compared with lower profile crimes such as burglary and
motor vehicle theft. It is suggested that perceptions of the former are more likely to be
influenced by high profile events and media coverage. Further evidence of differences
in perceptions come from the results of a question which showed that around half of
people surveyed (51%) thought they lived in a low crime area, 39 percent believed
that crime levels in their area were about average and only 11 percent thought crime
in their area was higher than average (Thorpe and Hall 2009: 98).
It has also been shown that people tend to over-estimate the risk of being a victim of
crimes such as assault, robbery, burglary and motor vehicle theft. For example, the
2008-09 BCS has found that 16 percent of respondents thought they were fairly or
very likely to be a victim of burglary compared to an actual risk of two percent
(Thorpe and Hall 2009: 99).
Studies have also looked at whether demographic factors affect variations in
knowledge. In a Home Office Study on public knowledge, little variation was found
in how poorly informed people were, although men did slightly better than woman
and those under 65 did better than those older than 65. Those who have had more
12
Part 2: Literature Review
contact with the criminal justice system had more knowledge, as do those who are
more interested in the criminal justice system (Chapman et al. 2002: 11).
Studies have also explored the extent to which public dissatisfaction with sentencing
can be traced to inaccurate perceptions about crime and criminal justice. Using
multivariate analysis of responses to the 1996 BCS, Hough and Roberts (1999: 18,
cited in Gelb 2006: 14) found that public misperceptions that were significantly
associated with a belief that sentences were too lenient included:

changes in national crime rates (those saying there was „a lot more crime‟
were most likely to think sentences were too soft);

estimated number of convicted muggers who were sent to prison (underestimators were most likely to think sentences were too soft);

the proportion of recorded crime involving violence (over-estimators were
most likely to think sentences were too soft);

estimated number of convicted burglars who were sent to prison (underestimators were most likely to think sentences were too soft).
Australian research findings are broadly consistent with those from other countries. In
Indermaur‟s Perth study, those who correctly estimated lower levels of violence
tended to favour less severe sentences (Indermaur 1987). Using the results of the
AuSSA 2003, Indermaur and Roberts (2003: 142-143) found that more than two
thirds of respondents reported that crime had increased over the past two years; more
than one third said it had increased „a lot‟ and only one in twenty or five percent of
respondents reported that it had decreased. Exploring the links between knowledge
and beliefs about criminal justice in the same survey in a later article, they found that
people who know more about crime rates are less punitive. However, they also noted
the importance of attitudes as well as knowledge, arguing that since confidence in the
courts is affected by emotive rather than instrumental concerns, public education must
address the symbolic and emotional issues that punitiveness reflects (Roberts and
Indermaur 2007: 61-62). Hough and Roberts (1999: 21, cited in Gelb 2006: 14) also
found that one-quarter of respondents thought that lenient sentencing was the most
important cause of rising crime rates, and almost half thought it was a major cause.
Despite significant evidence that factors affecting crimes rates lie largely outside the
reach of sentencers, the belief in a direct relationship between sentencing severity and
crime rates may lead many to blame judges for failing to control crime (Roberts and
Hough 2005: 48, cited in Gelb 2006: 14). Lack of knowledge about crime and the
criminal justice system is a significant factor in perpetuating public misperceptions
and misunderstanding (Gelb 2006: 14).
Those with high levels of fear of crime are more likely to be punitive (Sprott and
Doob 1997; Dowler 2003, cited in Roberts and Indermaur 2007: 58). The 1993
General Social Survey in Canada assessed the effect of people‟s prior victimisation
and fear (measured by feelings of safety walking alone at night and at home alone at
night) on crime. A total of 10,385 respondents aged 15 and over were randomly
sampled and interviewed by telephone. Of those who reported no fear, 71 percent felt
that sentences were too lenient. Of those with the highest levels of fear, 91 percent felt
this way. As fear increased, the proportion of people who thought that sentences were
too lenient also increased. The result held for victims and non-victims (Sprott and
Doob 1997: 281, cited in Gelb 2006: 19). However, Maruna and King (2004) estimate
that instrumental variables such as fear account for only four percent in explanatory
13
Jury Sentencing Survey
power of pro-community sanction attitudes over that provided by demographic
variables alone (cited in Roberts and Indermaur 2007: 58).
Several studies from different countries have shown that people with previous
experiences of crime victimisation are no more punitive than the general community.
For example, in the 1996 BCS, victims were given a crime vignette and asked to
impose a sentence. Comparing victims and non-victims showed that 55 percent of the
victims and 53 percent of the non-victims favoured imprisonment and even when
victims were victims of burglary – the same offence as committed by the offender in
the vignette - there was no difference between the groups in levels of punitiveness
(Hough and Roberts 1999: 21, cited in Gelb 2006). A survey in the United States of
1,056 adults found that crime victims were more supportive of
prevention/rehabilitation as opposed to punishment/enforcement than non-victims
(Hart, 2002, 19 cited in Gelb, 2006, 19). A UK survey of crime victims has shown
that victims of crime are more interested in prevention and the root causes of crime
than retribution (ICM Research 2006).
4. Demographic factors
An Australian Institute of Criminology (AIC) study found greater levels of
punitiveness among the less educated, males, lower income groups and the elderly,
but age differences were inconsistent (Walker, Collins and Wilson 1987: 3). Data
from the AuSSA survey shows that 12.8 percent of the variance in punitiveness is
explained by demographic factors although this is only part of the explanation.
Increased punitiveness is associated with being male, older and self-described as
working class, decreased punitiveness with more years of education (Roberts and
Indermaur 2007. For more on the demographic factors associated with punitiveness,
see Indermaur and Roberts 2005: 156). Older age groups were likely to favour
tougher punishment and women were likely to favour tougher punishment. (Note, this
finding was based on a punitiveness scale that was constructed by combining the three
survey questions on stiffer sentences, the death penalty and whether sentences should
reflect public opinion: Indermaur and Roberts 2005: 156). While the literature reveals
there are links between demographic factors and punitiveness, Roberts and Indermaur
(2007: 58) stated that they are „at best weak predictors of punitiveness‟. Their analysis
of the 2003 AuSSA data revealed that the number of years of education was the
strongest predictor of punitive attitudes. Only 12.8 percent of the variance in
punitiveness (as measured by questions asking respondents to agree or disagree with
the death penalty as the punishment for murder, that stiffer sentences should be given
and that judges should reflect public opinion) was explained by demographic factors.
5. The impact of information on punitiveness
In the light of the evidence that a lack of knowledge about crime and sentencing is
related to high levels of punitiveness as measured by a response to a general, abstract
question about sentencing, researchers have moved to ask questions that provide
much more information before asking for a response. Using case vignettes, either
fictional or based on actual cases, has been a popular strategy. A seminal series of 13
studies conducted by Doob and Roberts for the Canadian Department of Justice
demonstrated that while sentences described in the media were perceived by most
people as too lenient, those described in detail in court transcripts were mostly seen as
14
Part 2: Literature Review
appropriate. Study 9, for example, involved a comparison of extensive and detailed
coverage of a particular case in a newspaper with a detailed summary of court
transcripts. The study participants were 115 visitors to the Ontario Science Centre.
The study found that 63 percent of the media respondents felt the sentence was too
lenient compared with 19 percent of the transcript respondents. In fact 52 percent of
those reading the transcript felt that the sentence was too harsh (Doob and Roberts
1983: 31, cited in Gelb 2006: 18).
In addition to survey research, a number of studies have compared sentences imposed
by the judiciary with those imposed by lay members of the public. Diamond and
Stalans (1989) used vignettes in four moderately severe cases in which imprisonment
was a possibility but not inevitable. Respondents included 116 judges, jurors who
reported for jury duty and 55 university students. They were presented with detailed
information about each of the four cases including a pre-sentence hearing and video
of the sentencing hearing. They were told about the sentencing options available and
then completed a questionnaire indicating sentencing preferences. Judges‟ sentences
were as severe, or more severe than those of the lay respondents. No particular
differences between the offences in the four cases (a burglary, a robbery, a drug
offence and a wounding) were noted in this study. The study also looked at predictors
of sentence, such as appropriate goals of sentence, perceived seriousness of the
offence, prior record and demographic factors. The study asked lay respondents
whether Illinois judges were generally too severe, about right or too lenient. Sixty-six
percent thought they were too lenient. Why the standard polls question response
differs from the stimulus provided by individual cases was discussed and the answer
was found to lie in selective media reporting, social cognition and the way social
judgments are made. In asking a general question about court performance, the person
has to think about what the courts do and then evaluate the appropriateness of that
behaviour. For most members of the public, actual information about crime and
sentencing is incomplete and so knowledge is drawn instead from the media, personal
victimisation and the reports of others. The news media selectively reports crime and
sentencing and focuses on the violent and extreme rather than the ordinary case.
Research on social judgement suggests that the vividness of some media stories
would have a disproportionate impact on public perceptions, even if the media did
accurately portray the range of criminal behaviour (Kahneman and Tversky 1973
cited in Diamond and Stalans 1989: 87). When individuals respond to abstract
questions about judicial leniency, they attempt to recall prior cases and are influenced
by the severity of offences and offenders they recall. Tversky and Kahneman (1974
cited by Diamond and Stalans 1989: 87) label this biased process of recall, the
availability heuristic.
Lovegrove (2007) used judges and real cases in his Victorian study. He aimed to test
the populist view of judicial sentencing as lenient, and to trial a method of gauging
public opinion that addressed the need for the public to be aware of the principles and
factors relevant to sentence and to have a sense of the offender as a real person To this
end, Lovegrove provided participants with detailed information about the crime and
the offender, as well as information relevant to sentencing in individual cases, for
example the sanctions available and their cost. The exercise was presented to the
public as a consultation to address the perception that judges are out of touch.
Lovegrove arranged for two reserve and two recently retired County Court judges
who had each presided over a case to each explain it in detail to eight groups of
people. The 32 groups (each of about 15 people) were recruited from work places in
15
Jury Sentencing Survey
metropolitan and regional Victoria and presented with one of the four cases. In total,
471 members of the public participated. First, Lovegrove addressed each group for
about 70 minutes to explain relevant aspects of sentencing law and the sorts of issues
that judges are required to take into account when sentencing offenders. A brief
account was also given of the available sanctions and their dollar costs. In the second
session, the judge presented his sentencing judgment without revealing the sentence
he had imposed. Each group was told the maximum sentence and the average
sentence for the offence described in the case presented to them. They were then
asked to write down what sentence they thought was appropriate from a list of the
available sanctions. In three of the four cases where male offenders were imprisoned
by the judge (an armed robbery, a rape at knife point and the theft of over a million
dollars worth of goods by two employees), the median sentence imposed by the
members of the public was well below that imposed by the judge.
In the other case of a stabbing with intent to cause serious injury, the judge‟s sentence
fell just below the median sentence given by the public (Lovegrove 2007: 776). After
they were told of the sentence they were asked to rate the adequacy of the sentence on
a seven point scale ranging from „much too tough‟ to „much too soft‟ and to give their
views about the matters relevant to sentencing in the case. As would be expected, a
majority of those who had chosen a more severe sentence than the judge said that it
was too lenient, most strongly for the rape case (88%). But a significant percentage of
participants were prepared to defer to the sentence of the judge, particularly in the
case of the aggravated robbery (69 percent said it was too lenient, so 31 percent said it
was about right). This was even more pronounced for those who had chosen a more
lenient sentence with just 29 percent of respondents in the rape case saying it was too
harsh (compared with 67 percent in the aggravated robbery case). Lovegrove (2007:
778) notes that the responses were asymmetric for rape, intentionally causing serious
injury and theft: „It is the harsh who are apparently more certain of the correctness of
their view and less prepared to tolerate the court‟s sentence‟. He found this to
contradict the populist view of sentencing, which holds that the community „speaks
with one voice‟ and has firm views about what is an appropriate sentence.
Lovegrove (2007: 777-778) also suggests, contrary to the populist view of sentencing,
that the public relies on offender factors favouring leniency, not just on offence
seriousness. This conclusion was supported by: the expressed wishes of the
participants for treatment as well as custody for the offenders with personality
problems; the fact that participants imposed a suspended sentence on one offender for
whom prison would have been a special hardship; and the fact that participants cited
factors favouring leniency in their responses. Lovegrove concludes that people are
willing to give weight to mitigating factors even though the offending is serious
(Lovegrove 2007).
A recent Dutch study compared judges‟ choices of sentence with those of members of
the public in different surveys that enabled contrasts to be drawn between the public‟s
top-of-the-head opinions, their opinions based on the same case studies that were used
by the judges, and their opinions based on a short unbalanced newspaper version. It
found that providing the public with detailed information on a case had a strong
mitigating effect on severity but members of the public were still significantly harsher
than the judges. The authors conclude that there is indeed a gap between lay and
judicial punitiveness and that this gap could not be closed by additional or better
information (De Keijser, Koppen and Elffers 2007). As Lovegrove notes, while the
16
Part 2: Literature Review
case information given to the respondents was detailed, there was no information
about the criminal justice system or sentencing (Lovegrove 2007: 772).
The Scottish Justice 1 Committee research aimed to determine if citizens‟ views on
sentencing would change if they were presented with different types of information –
the study included survey interviews, focus groups and a type of deliberative poll. The
results confirmed that increased information decreases punitiveness but what is
particularly interesting about this study is the explanation it gives for why the public
tend to say that sentences are generally too lenient, even though, when faced with a
specific individual case, they select a sentence that is close to that likely to have been
imposed by the judge. Hutton explains this using Garland‟s distinction between
structuralist and individual accounts. Structuralist accounts deal with how the system
operates whereas a case scenario tries to find a just solution for an individual case
(Hutton 2005: 246). Public opinion about sentencing, it seems, is nuanced and
contradictory:
Punitive attitudes exist alongside more liberal views, perspective varies from the
local to the global and discussion about individual cases generates different
discourses from the practices of agencies and institutions (Hutton 2005, 246).
Hutton also utilises the idea of a narrative of insecurity: structural accounts of crime
are not based on accurate information about crime and risk but on an account which
expresses anxieties about broad patterns of social disorder (Hutton 2005: 251). The
punitive views expressed by the survey respondents are not less „real‟ than the views
expressed in the context of an individual case with better quality information (Hutton
2005: 246, 253-254).
Many writers have pointed out that any attempts to improve the levels of public
knowledge and the quality of the debate about crime and sentencing must be informed
by the need to be attuned to the emotional dimensions that construct public opinion.
„The real battle is not over facts and details but over morals and emotions‟ (Indermaur
and Hough 2002: 210). Freiberg (2001) speaks of the difference between effective
and affective justice (see also Ryan 2005) and others like Johnson (2009) have
explored the need for more research on the relationship between emotions and
punitiveness to augment the previous focus on cognitive and demographic factors
associated with the desire to punish criminals harshly.
6. Public views on the purposes of punishment
Surveys of public attitudes have sometimes examined people‟s perceptions of the
aims of sentencing (see Gelb 2006: 27 for some examples in the context of juvenile
crime). Hutton has reported on the results of the Scottish Justice 1 Committee‟s study
which asked survey respondents to select a sentence for an 18 year-old first offender
with a drug habit who committed a burglary and stole a video and then about the aims
of sentence in the case. There was more general support for rehabilitative and
reparative aims than incapacitation or simple retribution (Hutton 2005: 248). See also
Lovegrove (2007) discussed above.
Studies have also examined beliefs about the most effective way of controlling crime.
For example, a 2001 survey of 1,056 adults in the US found that 65 percent of the
adults surveyed favoured dealing with the root causes of crime and only 32 percent
preferred a punitive approach. Respondents reported that they strongly favoured
rehabilitation and re-entry programs over incapacitation as the best method of
17
Jury Sentencing Survey
ensuring public safety (Hart 2002: 4, cited in Gelb 2006: 25). There is also some
evidence that members of the public favour reparative options over incarceration
(Pranis and Umbreit 1992, cited in Roberts and Hough 2005: 139 and in turn by Gelb
2006: 28).
Indermaur‟s 1990 study of public attitudes to sentencing in Perth involved 410
members of the community and 17 judges and magistrates. Amongst other things it
looked at the most important purposes of sentencing for violent and property crimes.
For community members the most important purposes for violent offences were
incapacitation (37%), deterrence (24%) and retribution (23%). The responses from
judges were: deterrence (41%), incapacitation (24%), and retribution (18%). In the
case of property offences, the most important purposes for judges were rehabilitation
(71%) and deterrence (24%), whereas for community members, the most important
purposes were individual deterrence (49%) and rehabilitation (24%) (Indermaur 1990:
48-50; cited in Gelb 2006:31).
7. Confidence in the Courts and the Judiciary
There are some research findings which bear upon confidence in the judiciary. This is
relevant to the question in Questionnaire 1 and 2 which asks: How in touch do you
think judges are with public opinion about sentencing? A related question asked only
in Questionnaire 2 is: Should judges reflect public opinion about crimes when
sentencing criminals?
In Indermaur‟s 1990 Perth study, 57 percent of the public believed that public opinion
should be considered in all or most cases and 81 percent of sentencers believed this
(cited in Gelb 2006: 31). In the 2003 AuSSA most respondents (63%) felt that judges
should reflect public opinion in their sentencing decisions (Indermaur and Roberts
2003). They also found that 46 percent of respondents had „not very much‟
confidence in the courts.
Surveys which have asked whether judges are in touch include the BCS. The 1996
survey found that a substantial percentage (46%) believed judges to be „very out of
touch‟ and this was correlated with the assertion of judicial leniency (Hough and
Roberts 1998). The Scottish Justice 1 Committee study found 79 percent of
respondents thought judges were out of touch with what ordinary people think
(Hutton 2005: 247). The South Australian study used two statements to explore the
judiciary‟s relationship with the public. Participants were asked to respond to the
statement „the decisions of judges and magistrates reflect the views of the
community‟ – 41 percent agreed and 49 percent disagreed (South Australian Courts
Administration Authority 2007: 4, cited in Gelb 2008a: 4). The second statement was
„[it is] about time the courts caught up with the real world‟; 73 percent agreed with
this (Square Holes 2006, cited in Lovegrove 2006: 771). This suggests many
members of the public think that judges are out of touch with public perceptions and
that they do not make efforts to consider current public views on crime and justice.
Lovegrove uses these findings to support his argument for the need for judges to
consult with the community. He points out that judges appear remote and have a
different sense of justice because they differ from the community in terms of their
educational and social background and life experiences. They are not seen to consult
the community or to show an interest in what people think (Lovegrove 2007: 771).
18
Part 2: Literature Review
Surveys in a number of countries suggest there is a „crisis in public confidence‟
regarding criminal justice (Hough and Roberts 2004). Surveys comparing confidence
levels across criminal justice agencies have found that people have most confidence
in the police and least in the courts and prisons (Hough and Roberts 2004: 18). In the
2007 AuSSA survey, similar results were found: the Australian public has „greatest
confidence in the police, followed by courts and the least confidence in prisons‟
(Indermaur and Roberts 2009: 4, see also Roberts and Indermaur 2009: 20). When
asked why they have little confidence in the courts, respondents typically cite lenient
sentencing as the cause (Gelb 2008a: 3). A survey conducted by the South Australian
Courts Administration Authority found that 70 percent of respondents reported having
confidence in the state‟s courts, ranking courts fifth behind the police, the medical
profession, the public school system and the state government. But only half of all
respondents reported that they knew at least a little bit about the courts. So they had
an opinion about the courts knowing little about them (Square Holes 2006, cited in
Gelb 2008a: 3).
Roberts (2007) has discussed the difficulties of comparing confidence levels across
institutions. Reasons why it may be unreasonable to expect confidence levels in courts
to match confidence levels in institutions such as health and education, include the
different mandate the justice system has (Roberts 2007: 162), the greater influence of
ideology on confidence judgments in evaluation of courts than in health care delivery
(Roberts 2007: 163), misperceptions about crime trends (Roberts 2007: 164), and
differences in direct experience with public institutions – in the case of courts
information is usually indirect and filtered through the media (Roberts 2007: 165).
Similarly, higher confidence levels for the police than courts can be explained in
terms of different mandates and different levels of exposure (Roberts 2007: 173).
It should be noted that confidence levels may differ depending on whether the focus is
at the local level or nationally. A survey conducted for the Home Office in England
and Wales found that people were generally more confident with the way crime was
being dealt with locally than nationally (Page, Wake and Ames 2004).
While Sun and Wu (2006, 465 cited in Gelb 2008a, 3) found that people who have
had some recent experience with the courts hold more negative perceptions of the
courts, Benesh (2006, cited in Gelb 2008a: 3) has found that people with a high
control and low stake in the court system, such as jurors, are more supportive of the
court system than those without any experience of the court system, or those with
experience as defendants or plaintiffs who have a high stake and low control. The
results from the AuSSA 2007 survey indicate that Australians who had contact with
the courts over the previous 12 months „had higher levels of confidence in the courts
and were less likely to be in favour of tougher sentencing.‟ (Indermaur and Roberts
2009: 3).
8. Other questions measuring punitiveness
Questions on whether the public want to see stiffer sentences and support the death
penalty have been asked in the AuSSA since it began in 2003 (Gelb 2006: 35). In
2003, stiffer sentences were advocated by 70 percent of respondents with almost half
(47%) agreeing that the death penalty should be the punishment for murder. The
proportion advocating stiffer sentences has decreased since 1987 (Indermaur and
Roberts 2005).
19
Jury Sentencing Survey
The most recent results from the AuSSA 2007 survey confirm that:

the majority of respondents (58.4 percent) agreed that judges should reflect
the views of the public (Roberts and Indermaur 2009: 20);

four out of 10 respondents (43.5 percent) agreed with the statement that the
death penalty should be the punishment for murder, with just over one-third
(34.7%) disagreeing (Roberts and Indermaur 2009: 20); and

the majority of survey respondents (71.2 percent) agreed that „people who
break the law should be given stiffer sentences‟ with „only 6.6 percent
disagreeing and a further 22.2 percent neither agreeing nor disagreeing or
unable to choose‟ (Roberts and Indermaur 2009: 18).
9. Social psychological literature on attitude formation and the concept of
‘public opinion’
A number of researchers into public opinion refer to the literature on attitude
formation. Stalans (2002, cited in Hutton 2005: 245) argues that the social
psychological literature demonstrates that attitudes are dependent on a range of
factors including the structure of attitude in memory, and the ways in which the public
process information. Sparks (2002, cited in Hutton 2005: 245) has argued that survey
methods assume that punitiveness is something one can have more or less of, and
points out that this assumption conceals the contradictory views that appear when
more discursive approaches are used.
Green has noted the difference between mass public opinion and informed public
judgement (Green 2006: 132). Top-of-the-head responses to simple polling questions
represent mass public opinion, shallow, unconsidered views, as opposed to reflective
informed public judgment that emerges once people have engaged with an issue,
considered it from all sides, understood the choices it leads to, and accepted the full
consequences of the choices made.
In addition to the difference between opinion and judgment, writers have also
attempted to clarify the difference between „opinion‟ and „attitude‟. „Attitude is
traditionally conceptualised as a global, enduring orientation toward a general class of
stimuli, whereas an opinion is seen more situationally, pertaining to a specific issue‟
(Massen 1997, cited in Gelb 2008b: 3). Opinions are determined by attitudes which
are expressed positions or behavioural phenomena, while attitudes are the deeper
underlying motives for those behaviours.
Basing their views on the findings of the University of Cambridge Public Opinion
Project, Maruna and King (2004, cited in Gelb 2008b) suggest that lay people‟s
beliefs about why people commit crimes (attributional beliefs) may play a greater role
than actual experiences with victimisation (instrumental variables) in determining
attitudes to punishment such as support for community penalties. Attributional beliefs
were divided into two primary types: views that see crime as a choice (classical) or
views that see crime as a product of circumstances (situational). A second dimension
of attribution is a belief in a person‟s ability to change – „redeemability‟, and this may
override classical attributions. This was found to be the strongest predictor of support
for community penalties (Maruna and King 2004, cited in Gelb 2008b: 4). Measuring
attitude strength is another issue in survey research (Gelb 2008b: 4-5).
20
Part 2: Literature Review
10. Jury studies
While jury studies are now quite common, they have not, in the past, addressed
sentencing issues. One exception is the English Crown Court Study, undertaken for
the Royal Commission on Criminal Justice (Zander and Henderson 1993). This was a
general empirical study that explored how the criminal justice system works by
examining the views of the main actors in a sample of actual cases. It included a jury
questionnaire consisting of in excess of 80 questions in which the jury were asked just
one question in relation to the sentence, namely, „Was the sentence broadly as you
expected, based on the evidence in the case?‟ Only a minority stated the sentence was
higher (14%) or lower (23%) than they had expected (Zander and Henderson 1993:
223).
Jury studies are relevant to this project in terms of a comparison of response rates and
may also be relevant to the question of how representative jurors are of the general
population. Previous jury studies suggest that jurors are prepared to participate in
research projects relating to their jury service. In a jury study conducted for the New
Zealand Law Commission in 1998, an average of 54 percent of jurors in a total of 48
trials participated in interviews of more than an hour‟s duration about their
understanding of the law, the judge‟s directions, and their perceptions of the trial
process (Young, Cameron and Tinsley 1999). In a New South Wales study, the
response rate for completing questionnaires in sexual assault trials was 92 percent but
this dropped to between six to eight jurors per trial if they were allowed to take away
the questionnaire rather than complete it in the jury deliberation room (Cashmore and
Trimboli 2006). A study which examined facets of the quality and scope of the jury
experience in New South Wales, Victoria and South Australia achieved a response
rate from empanelled jurors of 75 percent (O‟Brien et al. 2008). Zander and
Henderson‟s English Crown Court jury survey was completed by 85 percent of jurors
from trials for which at least one juror responded (Zander and Henderson 1993). A
recent study of juror intimidation in Western Australia, which sent a 24-page survey
questionnaire to 2,954 jurors, achieved a response rate of 33% (975), with a further
454 consenting to an interview (Fordham 2009:44)
A possible flaw in using jurors to measure public opinion is that they may not be
representative of the general adult population, a bias that may be exacerbated by the
self-selection of jurors who are willing to participate in a jury sentencing survey. The
jury is promoted as being „representative‟ of community members. However, the
extent to which the modern jury is truly representative of the public in the sense of
being a cross-section of the community has been questioned. The wide range of
exemptions from jury service and the ease with which jurors are excused from service
are mentioned as reasons why a jury may not be truly representative (Victorian Law
Reform Commission 1997). Citizenship and English proficiency requirements mean
that jurors do not reflect the ethnic and cultural diversity of the community
(Australian Law Reform Commission 1992). Peremptory challenges further interfere
with the ability of jurors to be truly representative (Horan and Tait 2007; French
2007).
21
Jury Sentencing Survey
Early Australian studies showed significant age and gender discrepancies between
jurors and the general population (for example, Wilson and Brown 1973). However, a
recent study of civil juries in Victoria found that jurors were a fair cross-section of the
community in terms of gender and age, jurors from non-English speaking
backgrounds were marginally under-represented and university educated citizens
were over-represented (Horan and Tait 2007).
Psychological literature on decision-making
Other studies of relevance to this project are psychological studies which explore
decision making in relation to sentencing using lay persons and manipulating
variables such as victim impact statements and gender (Forsterlee et al. 2004) or
offender‟s awareness of risk (Feather, Boeckman and McKee 2001) and studies which
have explored the underlying motivations for punitiveness (Gaubatz 1994; Tyler and
Boeckmann 1997).
11. Mechanisms for improving public knowledge about crime and
sentencing
In recent years, judges and courts have become more pro-active about improving
public information about sentencing. One example is judges‟ involvement in „You Be
the Judge‟ sentencing workshops, in which members of the public participate in a
seminar with a mock trial component. Participants are asked to discuss what sentence
should be imposed (Warner 2007: 359, n 5). Similar events have been conducted by
Victoria‟s Sentencing Advisory Council. Another approach is to publish and
disseminate sentencing information. In 2007, the Judicial Conference of Australia
released a booklet entitled „Judge for Yourself: A Guide to Sentencing in Australia‟,
written by Peter Sallman with input from a steering committee and the Sentencing
Advisory Council of Victoria (Warner 2007: 359). The aim of the booklet was „to
educate the public and journalists in the face of what it believes is often unwarranted
criticism‟ (Debelle J, cited in Warner 2007: 359). Publicising the booklet‟s launch in
September 2007, Debelle J, the Chairman of the Judicial Conference, was reported as
saying it was intended to help the public understand the process of sentencing and to
„weigh criticism in the media. It is also provided for the purpose of educating
journalists to be more temperate in their criticism‟. Another publication is planned to
be „aimed at a more informed audience‟ (Debelle J, cited in Warner 2007: 359).
An English study commissioned by the Home Office used three methods of
presentation of information (a booklet, a seminar, and a video) to test the impact of
the information on knowledge, confidence, and attitudes (Chapman, Mirrlees-Black
and Brown 2002). The study found that providing simple factual information about
crime and sentencing can improve public knowledge of these matters in the short term
at least, and that it has an impact on attitudes and confidence in the criminal justice
system. After receiving the information, participants were less worried about being
victims of crime, and less likely to say sentencing is currently too lenient. Each of the
three information formats tested produced similar improvements in knowledge;
although the improvements were significant, only one person got all 11 questions
right on the follow-up interview (Chapman et al. 2002: 11-14). The booklet was found
to be the most cost-effective of the formats tested and it also reached the widest crosssection of people. Despite financial incentives, participation in the seminar was very
low, and there was also a poor participation rate with the video format. The authors
22
Part 2: Literature Review
noted that the marked improvement in attitudes to the criminal justice system may not
be due to improved knowledge about crime and criminal justice but because of the act
of engaging in the exercise (Chapman et al. 2002: 50).
A later study looked at the impact of a 20-page booklet, Catching up with crime and
sentencing on a sub-section of respondents from the British Crime Survey, who were
re-interviewed about two weeks after receiving the booklet. It found that a quarter of
respondents said that it had changed their views; reading or flicking through the
booklet improved knowledge but knowledge of the proportion of rapists or burglars
sent to prison was not improved. Confidence increased in all aspects of the criminal
justice system for those who looked at the booklet but some increases were as a result
of engaging in the topic through taking part in the interviews (Salisbury 2004).
Improving the level of public knowledge about crime and punishment has been
suggested as an obvious remedy to combat „penal populism‟, defined as „allowing the
electoral advantage of a policy to take precedence over its penal effectiveness‟
(Roberts et al. 2003: 5). Roberts et al. have suggested that, governments must provide
much clearer information on crime trends and the „going rate‟ of sentences for
specific sorts of crimes; that audiences need to be identified and targeted; and new
technology such as interactive websites used to convey the information (Roberts et al.
2003: 168-174).
12. The limits of public education using information
Not all are convinced about the value of attempting to educate the public by providing
accessible information. For example, Green (2006) has argued that the kinds of public
education programs embraced by the Home Office, such as distributing booklets or
videos are insufficiently bold to make a significant and lasting impact on public
knowledge and attitudes. These programs are inherently flawed because they do not
help the public work through the ambivalent attitudes that crime and punishment
issues often produce, to enable the development of more considered views.
Information is a necessary condition for attitude change but it is not sufficient:
[T]hese educative approaches do not generate the deliberation and dialogue
needed to produce a durable public judgment. What is required is the
development of informed preferences for which citizens take responsibility and
which endure over time in the face of emotive rhetoric and the next high-profile
tragedy. Instead, these approaches engage the public on a technical and
informational level – an expert‟s framework – disallowing the release of
Yankelovich‟s “bees in bonnets” before new information is introduced (Green
2006: 146).
For Green, the answer lies in the deliberative poll. Participants, a large stratified
random sample of the public, are provided with balanced briefing materials, then
gather for a weekend to hear presentations, engage with experts and debate among
themselves. In this way, informed public opinion can be obtained by first enabling its
creation.
Similarly, Maruna and King have cautioned that „public education will help, but is no
panacea‟ (2004: 101). Consistently with the Home Office study discussed above,
surveys invariably find that providing respondents with more information about
sentencing options and the offenders themselves has an immediate impact on reducing
punitive tendencies. However, Maruna and King give a number of reasons for their
23
Jury Sentencing Survey
reservations (2004: 101). First, most research on the impact of education on attitudes
shows only very short-term effects or at least that the duration of the effects is
unknown. Second, much of the research is plagued by a „Hawthorne effect‟, namely
„participants may modify their views on follow-up surveys simply because that is
what they are supposed to do‟. Finally, the practicality of introducing many of these
educational efforts on a large scale is doubtful. The effectiveness of deliberative
polling as a means of changing deep-seated attitudes is questioned, as is the
practicality of educating the public in general through this method. They argue that
attitudes have an emotional dimension as well as a factual one, and suggest that when
attitudes to crime, sentencing and penalties are not merely based on information
deficits, they are not easily altered (Maruna and King 2004: 102) Others have also
argued for the need to address the emotional attitudes to crime and justice (Freiberg
2001). In conclusion, Maruna and King state:
Schemes to educate and inform the public about the nuances of sentencing, the
“facts” about crime, and so forth are noble, well-meaning efforts, but unlikely to
have more than marginal impact on either public understanding of crime issues or
punitive, prison-centric attitudes (2004: 102).
However, they note that the most promising findings about the impact of education is
in the context of active participation by citizens in the criminal justice process, such
as serving on a jury or participating in restorative justice work. Research suggests
active participation increases satisfaction with the criminal justice system and
decreases punitiveness (Maruna and King 2004: 102). This suggests that using jurors
as a means of educating the public has some potential. Moreover, they could be used
as a source of informed public opinion.
24
PART 3
RESULTS
A. STAGE ONE
1. Information on the cases in the sample
Quantitative information
The study ran for two years and included 162 trials. All trials in which there was a
verdict of guilty on at least one count were included. This section includes some basic
descriptive data on the cases in the sample. The type of crime for which guilty
verdicts were received is shown in Table 1 below. Where there were multiple counts,
the most serious crime is recorded. The Table shows the information for all 162 trials
and the 138 trials for which at least one response was received. It shows that the
distribution of offence type was very similar for trials with some participation and for
all 162 trials in the study.
Table 1: Type of crime
Sex
Violence
Drugs
Property
Other
Culpable
driving
%
N
a
%
N
%
N
%
N
%
N
%
N
162 trials
18.5
(30)
35.2
(57)
22.2
(36)
15.4
(25)
6.8
(11)
1.9
(3)
138 trials
17.4
(24)
36.2
(50)
23.2
(32)
14.5
(29)
7.2
(10)
1.4
(2)
a
Other includes: fabricating evidence (1); perverting the course of justice (3); making a false declaration (1); corrupting witness
(1); attempting to interfere with a witness (2); compounding a crime (1); conspiracy (1) and harbouring (1).
The 182 defendants were predominantly male (86%) and 90 percent of trials involved
only the one defendant. Of the 138 trials with at least one response, 12 involved two
defendants, and one trial had three defendants. Where there were co-offenders,
respondents were asked to select a sentence for each offender.
Table 2 shows the range of penalties imposed by the judge in the case tried. Again,
data is shown for all 162 trials and the 138 juror response trials. All jointly charged
defendants are included so the total in the rows is greater than the number of trials.
Table 2: Most serious sentencing outcome imposed by judge (for all offenders)
Custodial
sentence
162 trials,
182
offenders
138 trials,
153
offenders
Wholly
suspended
sentence
Community
service
order
Probation
order
Fine
Disqualified
from holding
motor vehicle
licence
%
N
%
N
%
N
%
N
%
N
%
N
75
(136)
18
(33)
0.5
(1)
1.6
(3)
3.8
(7)
0.5
73
(112)
18.9
(29)
0.6
(1)
1.9
(3)
3.9
(6)
0.6
Proceeding
adjourned
%
N
(1)
0.5
(1)
(1)
0.6
(1)
25
Jury Sentencing Survey
In all 162 trials in the study period, 93 percent of the sentences imposed by the judge
were custodial, and most were immediate sentences of imprisonment. Eighteen
percent of all sentences were wholly suspended prison sentences and just seven
percent were non-custodial. This is a smaller proportion of non-custodial sentences
and a larger proportion of immediate custodial sentences than one would expect from
a sample of all offenders sentenced in the Supreme Court. In 2002-2004, noncustodial orders comprised 14 percent of all sentencing dispositions in the Supreme
Court of Tasmania and wholly suspended sentences comprised 29 percent (Bartels
2008: 167). This suggests the cases in the study are more serious than offenders
generally dealt with by the Supreme Court. Table 2 also shows that offenders in trials
for which responses were received attracted a similar range of sentences as all
offenders in the study.
Appeals
In total there were 11 sentencing appeals in the cases in the study, two of which were
successful – with the sentence being increased in one and reduced in the other. The
sentences were not adjusted in the database. There was no participation in one of the
cases and in the other nine jurors participated in Stage 1, three selecting a more severe
sentence than the judge and six a more lenient sentence. On appeal the sentence was
increased with the result that four of the jurors‟ sentences were more lenient and five
more severe (Hales v Tasmania [2009] TASSC 100). In Stage 2, all seven participants
said the judge‟s sentence was very appropriate. It follows that the results of
sentencing appeals impact little on the comparison between the Court‟s sentences and
those selected by the jurors.
Were sentencing submissions heard by respondents?
As explained in Part 1, in some cases sentencing submissions are adjourned. They
may be adjourned until another day or adjourned for an hour or less. Sentencing
submissions are made by the prosecution and the defence. The prosecution‟s address
can highlight the facts in relation to any aggravating features of the offence and may
include a victim impact statement which may be read by the prosecution counsel or
by the victim. The prosecution will also supply the court with the criminal record of
the defendant and information on whether the defendant was on parole, probation or
bail at the time of the offence. The prosecution is permitted to make suggestions in
relation to the type of sentence that should be imposed. Defence counsel has the duty
to make a plea in mitigation. This can include facts in relation to the offence and the
offender. The defence may challenge the factual basis of a sentence put forward by
the prosecution and argue for a version of the facts consistent with the verdict that is
more favourable to the defendant. The defendant‟s social background and
employment history may be explained and submissions made in relation to prospects
of rehabilitation.
In this study, if sentencing submissions were adjourned, a transcript was sent to the
respondent in Stage 2 with Questionnaire 2 and the information booklet. In
Questionnaire 1 jurors were asked whether they had listened to the sentencing
submissions. Seventy five percent of Stage 1 respondents said they had stayed to
listen to the sentencing submissions. In fact, sentencing submissions were only
adjourned or adjourned in part in 13 percent of cases, so it is possible that some of
those who said they were not present were in fact present. If submissions were
adjourned the participation rate in the study was slightly less.
26
Part 3: Results
2. Are jurors willing to be a source of public opinion on sentencing?
The current study was demanding of participants. It required jurors who have already
been inconvenienced by jury service to remain in court to listen to the sentencing
submissions (if they were heard immediately after verdict) before completing
Questionnaire 1. While sentencing submissions generally last no more than 30 or 40
minutes, they can be longer if there is a factual dispute and evidence is called. To
participate in Stage 2, jurors were required to read the information booklet, the
sentencing remarks and to fill in another Questionnaire. If the sentencing submissions
had been adjourned they also had a transcript of the sentencing submissions to read.
Stage 3 participation required a face-to-face interview. No incentives were offered to
jurors to encourage participation. The study was also demanding of judges, associates
and court personnel as the study ran for two years and study fatigue was a possibility.
Court personnel were required to ensure they had an adequate stock of questionnaires,
consent forms and information sheets for distribution and were asked to contact the
research team for replacements. Associates were asked to supply information about
length of trial, time of verdict etc.
Table 3 shows a response rate of 36 percent for all of Stage 1 and the response rate for
the first 51 trials in the study (42%). This compares with previous jury studies which
have achieved response rates from 33 percent (Fordham 2009) to 92 percent
(Cashmore and Trimboli 2006).
Table 3: Juror response rate
No. of trials
No. of jurors asked to participate
All 162
First 51
1944
612
Response rate
%
36
42
N
(698)
(257)
Agreed to Stage 2
%
88
90
N
(614)
(231)
The response rate for this study was impacted by a number of factors. No jurors
participated in 24 trials. In the other 138 trials (85 percent of trials) at least one juror
participated. Participation in these trials ranged from 1-12 jurors. The median
participation rate for juries with at least one response was five. As participation rates
seemed to decline after the first fifty or so trials, the response rate for the first 51 trials
was also calculated. This shows that there was a fall-off in participation in Stage 1
after the first 51 trials. As Tables 4, 5, 6, 7 and 8 indicate, response rates also varied
by judge, place of trial, time of verdict, length of trial and type of offence.
Juror participation relied on the judge‟s invitation to jurors to participate in the study
after the verdict was announced. While there is no reason to believe that any judge
forgot to remind the jury of the study and invite them to participate, the different
response rates shown in Table 4 suggests different approaches by the judge in issuing
the invitation to jurors to participate in the study may have affected the response rate.
Excluding Judge 1, who retired soon after the study commenced, the proportion of
jurors responding varied from 27% (Judge 7) to 44% (Judge 2) and the proportion of
trials with a nil response varied from 35% (Judge 6 and Judge 7) to 3% (Judge 2).
27
Jury Sentencing Survey
Table 4: Juror participation by trial judge
Judge
1
Judge
2
Judge
3
Judge
4
Judge
5
Judge
6
Judge
7
Total
0
1
4
2
3
6
8
24
1
32
36
28
15
11
15
138
No. of Respondents
11
173
186
137
61
56
74
698
Participation rate
92%
44%
39%
38%
28%
28%
27%
36%
Trials where no
juror participated
Trials where jurors
participated
The data were analysed to see if there was a difference in the response rate between
the three cities where trials were heard, Hobart, Launceston and Burnie. As
anticipated, (see Table 5) Hobart trials provided the highest proportion of
respondents. However, it also provided the best response rate. The response rate was
poor in Burnie, a small city on the North West Coast of Tasmania. This was predicted
by court staff as the facilities for jurors were said to be poor at the Burnie Supreme
Court. It is also possible that the response rate in Burnie was affected by the fact the
research team had less personal contact with court staff during the study. In Hobart,
court staff were particularly supportive of the project and encouraged juror
participation. Moreover, it is possible that jurors may be more inhibited in
participating in a study in smaller cities such as Burnie where the chances of seeing
the defendant or his or her family after the trial are higher.
Table 5: Juror participation by place
Place of trial
Hobart
Launceston
Burnie
Total
Total No. of trials
75
52
35
162
900
624
420
1944
412
188
98
698
59
27
14
100
46
30
23
36
3
13
8
24
4
25
23
15
Total No. of
jurors
No. of
participants
Response rate
%
Local response
rate %
No. trials with nil
response rate
Trials with nil
response rate %
It was also hypothesised that the type of offence could make a difference to response
rates. Jurors may be more interested in certain kinds of trials and/or they may be more
interested in stating their views on sentencing in relation to particular types of crime.
This was tested by looking at the types of offence for which there was a very good
response rate. Table 6 shows that there were no significant differences in the
participation rates between offence categories. Juror participation rates for sex,
violence and property trials were very similar, and the rates for drug trials only
slightly lower. As there were only three trials for culpable driving, little can be drawn
from these figures. Examining the measures of participation rate by type of crime
suggests that even or median response rates were more likely for violent and drug
offences.
28
Part 3: Results
Table 6: Juror participation by type of offence
Type of offence
Sex
Violence
Drugs
Property
Culpable
Driving
Other
Total
No. of trials
30
57
36
25
3
11
162
Total no. of jurors
360
684
432
300
36
132
1944
No. of participants
128
253
143
113
8
53
698
Offence response rate
(%)
36
37
33
38
22
40
36
% with nil response
20
12
11
20
33
9
15
All 12 jurors participated in the cases of:

Ridley (where a woman was convicted of dishonestly acquiring a financial
advantage when she applied for six loans and failed to disclose the extent to
which she was in debt);

Rogers (both offenders convicted of assault after punching and kicking the
complainant and then driving a motor vehicle at the complaint); and

Pannala (19 instances of computer fraud against his employer, involving
$1,219,539.05).
11 jurors participated in the cases of:

Martin (attempting to interfere with a witness: a Doctor paid a violent client,
to harm a former lover/client and her family, so she would not lodge a
professional misconduct complaint); and

Morley (whilst in a jealous rage, Morley assaulted his ex defacto by pushing
her off the bed).
As Table 7 suggests, the length of trial had little impact on response rate. Shorter
trials had the poorest rather than the best response rate although these trials were less
likely to have no response. Nor did length of deliberation have much impact on the
response rate although trials where jurors deliberated for less than two hours had a
better response rate than longer trials (Table 8).
Table 7: Juror participation by length of trial
Short trial
(1-2 days)
Medium trial
(3-5 days)
Long trial
(5 + days)
Total
No. of trials
88
51
21
160
Total no. of jurors
1056
612
252
1920
236
90
689
39
36
36
7
5
24
14
24
15
No. of
363
participants
Jurors response
34
rate %
No. trials where
12
no participation
Trials with nil
14
response %
a
Data on two trials (nine jurors) missing.
a
29
Jury Sentencing Survey
Table 8: Juror participation by length of deliberation
No. of trials
Total no. of
jurors
No. of
participants
Response rate %
No. trials where
no participation
Trials with nil
response %
a
Short deliberation (2
hours or less)
Medium deliberation
(3 hours or less)
Long deliberation
(more than 3 hours)
Total
51
52
57
160a
612
624
684
1920
239
231
209
679
39
37
31
35
7
8
9
24
14
15
16
15
Data on two trials (nine jurors) missing.
In addition to the data presented above, we gathered comments from the judges‟
associates on differential response rates including (some, but not all) nil response
trials. These comments suggest that late verdicts, long trials, lack of interest, fear and
intimidation may be factors. In one case, the sentencing submissions were adjourned
for 45 minutes but no jurors returned to listen to the sentencing submissions. In
another they were adjourned until after lunch. Although it seems that almost all jurors
returned to listen to the submissions in this case, no questionnaires were returned. We
surmise that in both of these cases the questionnaires were not offered to juries.
A high proportion of jurors who completed Stage 1 (88%) agreed to participate in
Stage 2. At the end of Questionnaire 1, jurors were invited to explain why they
declined to participate further if in fact they had declined. Almost a third of those who
declined to participate in Stage 2 responded to this question. There are four themes in
these responses. First, the juror was too busy, could not be bothered or felt they had
done their duty. Secondly, a lack of confidence in their ability to say anything useful
(the questions on crime and sentencing trends may have put them off). Thirdly, a fear
of loss of anonymity or breach of privacy by disclosing their name (this was
necessary for participation in Stage 2). And finally, two respondents expressed
disappointment/dissatisfaction with the criminal justice system.
3. How representative are jurors of the general population?
As discussed in the literature review, one problem with using jurors to measure public
opinion is that they may not be representative of the general adult population. In
Tasmania, the Juries Act 2003 (Tas), which commenced on 1 January 2006, has
drastically reduced the number of occupations that render a person ineligible for jury
service and tightened the grounds for application for excuse. However, the jury pool
is unlikely to be representative of the general community in terms of ethnic
background because eligibility for jury service depends both on enrolment on the
Electoral roll (Juries Act 2003 (Tas) s 6(1)), which in turn depends on citizenship
(Electoral Act 2004 (Tas) s 31; Commonwealth Electoral Act 1918 (Cth) s 93) and an
adequate ability to communicate in or understand English (Juries Act 2003 (Tas) s
68(3) Schedule 2, item 10). Successful applications for excuse and deferral could also
reduce the representativeness of the jury pool.
30
Part 3: Results
There are no studies of the representativeness of Tasmanian juries and the Supreme
Court does not collect juror demographic data. However, as demographic data was
collected from Questionnaire 1 respondents, it was possible to compare the
respondents with the jury eligible population in terms of age to assess how
representative they are of the general population.
This comparison is outlined in Table 9 below. In terms of gender and age, the jurors
who responded to Questionnaire 1 are reasonably representative of the jury eligible
population. Females were a little over-represented, as were persons in the 45 to 64 age
group. Not surprisingly, people 65 and over are under-represented because persons
over the age of 70 can apply to be excused. The Questionnaire 1 jurors were also
more educated, with jurors more than twice as likely to hold a degree qualification.
Data on place of birth was collected from respondents and was regarded as an
acceptable proxy for ethnic background. As expected, because of citizenship and
language requirements, Australian born respondents were over-represented in the jury
respondents (91 percent of juror respondents were Australian born compared with 83
percent of the Tasmanian population). Our juror respondents were also less likely to
be unpartnered than the general jury-aged population, and given the underrepresentation of people 65 and over, they were also less likely to be widowed and not
currently partnered.
An accurate comparison cannot be made for income levels because of differences in
data collection, but an approximate comparison (not reported here) indicates that
jurors are less likely to have a lower income than the general jury-aged Tasmanian
population. As a result, it is not surprising that the juror respondents were more likely
to be in full-time employment than the jury-aged Tasmanian population and less
likely to be unemployed. In terms of current occupation, the juror respondents were
very representative of the general jury-aged Tasmanian population.
Overall, while some differences between the juror participants and the jury eligible
population were found, the juror respondents provided a basic, but not mirror,
reflection of the broader Tasmanian population. Hence, while results from this study
can be fully generalised, they may also be cautiously viewed, where relevant, as an
indication of broader public sentiment
31
Jury Sentencing Survey
Table 9: Study jurors and Tasmanian jury eligible population
Socio-demographic variable
Juror respondents
%
Jury aged population
%
Female
53.4
51.8
Male
46.6
48.2
18-24 years
10.1
11.5
25-44 years
33.2
33.4
45-64 years
48.3
35.5
65+ years
8.4
19.7
Bachelor's degree or above
24.4
11.9
Not stated
0.7
12.3
91
83.2
Single/never married
17.1
25
Married or partnered
Separated/divorced and not currently
partnered
Widowed and not currently partnered
71.9
(married only) 53.8
9.4
13.7
1.6
7.5
Full time
48.9
34
Gender: N = 695
Age group: N = 690
Education level: N = 698
Country of birth: N = 694
Australia
Marital status: N = 690
Employment status: N = 690
Part time/casual
20.2
19
Employed away from work
1.4
2
Unemployed looking for work
1.6
3
Not in the labour force
26.8
37
Missing data
1.1
5
Current Occupation: N = 538
Managers, professionals and
paraprofessionals
Trades
41.4
42.2
11
13.9
14.4
Clerical
14.5
Production, labourers, elementary
29.9
27.5
clerical
Not stated or inadequately described
3.7
1.4
a
Source: Derived from ABS, 2006 Census of Population and Housing, Tasmania (State) Tables
32
a
Part 3: Results
4. From where do jurors source crime and sentencing information?
To provide a baseline, respondents were asked about where they sourced their crime
and sentencing information. The question was phrased as follows: „People get
information about crime and sentencing from a variety of sources. Please indicate
whether the following are a major source, a minor source or not a source.‟ This was
followed by seven possibilities. The responses are shown in Figure 1.
The Australian Survey of Social Attitudes (AuSSA 2007) found most Australians rely
on broadcast and print media as their most important source of news about crime and
criminal justice. Almost eight out of ten respondents rated each of TV, radio and
newspapers as fairly or very important (Roberts and Indermaur 2009: 9). Similarly,
for jurors, television, newspapers and radio were the most important source of
information about crime and sentencing. Internet sites and family and friends were
least likely to be an important source for both jurors and AuSSA respondents.
Figure 1: Sources of information about crime and sentencing
80.0%
Percentage
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Internet
sites
N=655
ABC/SBS
TV
N=664
Commercial
TV
N=660
Radio
N=663
Newspaper
s
N=686
Talkback
radio
N=648
Friends and
Family
N=664
Major source
20.3%
50.2%
43.3%
43.6%
70.4%
10.2%
15.1%
Minor Source
33.6%
37.7%
43.2%
44.5%
27.6%
36.1%
59.5%
Not a source
46.1%
12.2%
13.5%
11.9%
2.0%
53.7%
25.5%
N = 648-686
5. Crime victimisation
Questionnaire 1 asked respondents if they had ever been a victim of a crime that was
reported to the police. Forty one percent of participants responded positively. In
Figure 2, the responses are classified into six categories: sex, violence, drugs,
property, culpable driving and other. On the basis that 13 percent of recorded offences
in Tasmania are violent offences and 82 percent are property offences, our jurors
would appear to be representative of the general population in terms of type of
victimisation.
33
Jury Sentencing Survey
Figure 2: Type of crime committed against victims
Sex 2%
Violence 14%
Drugs 0% when rounded
Property 82%
Culpable driving 0% when rounded
Other 2%
6. Juror’s proposed sentence compared with judge’s sentence
The third question in Questionnaire 1 asked respondents to indicate what sentence
they thought the offender should receive. Questionnaire 1 was completed before the
judge imposed the sentence (or at least before the juror knew of the sentence). The
question included a menu of sentencing options to ensure that respondents were aware
of the alternatives available and did not focus unduly on the better-known sentencing
options such as imprisonment and fines. This was done in the light of the research
finding that, when given sentencing vignettes, respondents who had to choose a
sentence without the benefit of a menu were significantly more likely to favour
imprisonment than those who had a menu (Hough and Roberts 1999: 19). In the menu
of options discharging the offender without a conviction was not listed as this option
is rarely used in the Supreme Court. Nevertheless, it was imposed in one case in the
study.
To compare the juror‟s sentence choice with the judge‟s sentence we constructed an
imputed variable „Stage 1 comparative sentence severity‟. This unique variable was
used to gauge punitiveness in the study. The following rules were used to code the
variable. Sentencing options were ranked in ascending order of severity:

Conviction recorded and discharge

Fine

Probation Order

Rehabilitation Program Order

Community Service Order

Wholly suspended sentence of imprisonment

Imprisonment
Under this categorisation, if the juror selected a wholly suspended sentence (of
whatever length) this was counted as less severe than an immediate term of
imprisonment (of whatever length). If the juror selected more than one sentencing
order, the most severe was compared with the judge‟s (most severe) sentencing order.
If the juror‟s sentencing option was the same as the judge‟s, then severity depended
on the length, number or amount of the order. In the case of partly suspended
34
Part 3: Results
sentences, severity depended on the immediate term served. Because the jurors were
not given the option of a discharge without a conviction, in the one case this was
ordered (Tasmania v Cooper, 12 September 2007) it was coded as the most lenient
sentencing option listed, namely a „conviction recorded and discharge‟. In the one
case where a sentence to the rising of the court was ordered (Tasmania v Hume, 5
February 2009), this was coded as a sentence of imprisonment but where jurors
selected a conviction recorded only or a wholly suspended sentence this was recorded
as the same as the judge. Where a sentence of imprisonment was imposed but
backdated, this was still counted as a sentence of imprisonment and whether it was
coded as more or less severe than the juror‟s sentence depended on the term of
imprisonment from the date of the sentence.
Comparing the juror‟s sentence with the judge‟s sentence, as detailed in Figure 3,
showed that 52 percent chose a more lenient sentence than the judge and 44 percent
chose a more severe sentence. Figure 3 cross-tabulates the responses by type of crime.
The low „Same‟ figures are an artefact of the wide menu of alternatives available.
Figure 3: Judge and juror’s sentence compared by type of Offence
Percentage
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Total
N=706
Sex
n=118
Violence
n=278
Drugs
n=137
Property
n=109
Culpable
driving
n=11
Other
offences
n=53
Less severe
52%
48%
49%
48%
68%
82%
43%
Same
4%
4%
5%
1%
2%
0%
6%
More severe
44%
48%
46%
50%
30%
18%
51%
N = 706. This figure includes all jurors’ responses for all offenders
These results indicate that:

Jurors were more likely to be less severe than the judge than more severe for
all offence types except for drugs, where 50 percent were more severe and 49
percent less severe. For „other offences‟ they were also more likely to be
more severe.

Jurors were most likely to be less severe than the judge for property offences
and culpable driving. However, for culpable driving there were just three
trials and 11 respondents, so it is difficult to draw conclusions in relation to
this offence category.

Jurors were evenly split between more and less severe sentences than the
judge for sex, violence and drug offences and were least likely be less severe
than the judge for „other offences‟.
35
Jury Sentencing Survey
A bivariate analysis of the severity of the proposed sentence compared to judge‟s
sentence (Stage 1 comparative sentence severity) with offender characteristics found
that:

females were less likely than males to select a more lenient sentence (49-56
%);

those in the 18-24 age group were more severe than those in the 45-64 age
group;

no relationship was found between marital status, income level or education
level and comparative sentence severity.
As detailed earlier, one quarter of respondents said they had not listened to the
sentencing submissions before indicating their choice of sentence and therefore were
not informed of victim impact or criminal history information. Nor had they listened
to the plea in mitigation. Comparing the Stage 1 comparative sentence variable of
those who were present for the sentencing submissions with those who were not,
showed that 47 percent of those present chose a more severe sentence than the judge
compared with 44 percent of those who said they were not present – statistically, a
non significant difference.
7. Jurors’ knowledge of crime trends and sentencing patterns
Four questions tested participant‟s knowledge of crime and sentencing. The first was,
‘Do you think that recorded crime over the last 5 years has increased a lot, increased a
little, stayed about the same, decreased a little, or decreased a lot?‟ More than a third
thought it had increased a little and more than a quarter thought it had increased a lot.
Jurors‟ responses to this question are juxtaposed below with the pattern of recorded
crime in Tasmania between 1981 and 2006.
Table 10: Jurors’ perception of overall recorded crime trends
Increased a
lot
%
27
Increased a
little
%
36
Stayed the
same
%
15
Decreased a
little
%
6
Decreased a
lot %
Don’t know
%
1
15
As shown for the last five years displayed in Figure 4 there was a decrease in the
recorded crime rate. This downward trend has continued to present (Department of
Police and Emergency Management 2009: 79, 81). This trend is also evident
nationally with declining crime rates for property and violent crimes (for example,
robbery and homicide) (Australian Institute of Criminology 2009, Facts and Figures
2008). Therefore, contrary to what the overwhelming majority of respondents
indicated they believed, recorded crime decreased in the last five years.
36
Part 3: Results
Figure 4: Rate for recorded crime, Tasmania, 1981-82 to 2005-06.
16000
14000
12000
10000
8000
6000
4000
2000
0
Offences reported per 100 000 population
Compiled from recorded crime data in Tasmania Police Annual Reports and ABS Population by Age and Sex, cat no
3201.0; note: recorded crime data covers offences collected for national statistics for the ABS and does not include
all criminal offences, notable exclusions are offences with no identifiable victim - so drug offences and driving
offences are not included.
In relation to crime trends for specific crimes recorded in Tasmania over the last five
years, jurors were asked whether the following crimes had become more or less
common: burglary, robbery, rape, motor vehicle theft and murder. Again, the
responses shown in Table 11 do not align with current data with less than 10 percent
of jurors accurately estimating the direction of recorded crime across most categories.
In relation to burglary, respondents were most likely to say that it had stayed the
same, although the number of recorded burglaries of houses and commercial premises
has declined since 1998 (Department of Police and Emergency Management 2009).
The number of robberies peaked in 1998-1999, and since 2005-2006 there have been
fluctuations but the trend has been more downwards than upwards or stable. For rape
the trend is less clear, but the most accurate response is that it has stayed the same
(Weatherburn and Indermaur 2004: 4). In relation to motor vehicle theft, most
respondents said that it had become more common although the general trend for
motor vehicle theft is downwards. The number of motor vehicle thefts fell from 2,375
in 2002-2003 (Department of Police and Public Safety 2004, Annual Report 20032004) to 1,618 in 2006-2007 and 1,382 in 2008-2009 (Department of Police and
Emergency Management, 2009).
Table 11: Jurors’ perceptions of crime trends
Juror’s perception of crime trends
Crime Type
More common
%
Stayed the same
%
Less common
%
Don’t know
%
Burglary (N = 690)
57
29
6
8
Robbery (N = 687)
57
31
4
8
Rape (N = 681)
17
47
15
20
Motor vehicle theft
57
27
7
9
This finding of poor knowledge about crime trends is broadly consistent with other
Australian research and with international findings. Most members of the public think
crime rates are rising even when they have been falling over a number of years (Gelb
37
Jury Sentencing Survey
2006: 13, 30). For example, using AusSSA 2003 data, Indermaur and Roberts (2005:
143) found that 39 percent of respondents thought that crime rates had increased a lot
over the last two years, 31 percent thought it had increased a little, 21 percent thought
it had stayed the same, 4 percent thought it had decreased a little, and 1 percent
thought it had decreased a lot. The state-by-state breakdown indicated that for most
states the most common response was „increased a lot‟. Consistent with our results,
for Tasmania and Victoria it was „increased a little‟. AuSSA 2007 results are similar
but report even less respondents (2.9%) correctly identifying that crime rates had
reduced (Roberts and Indermaur 2009: 9). Weatherburn and Indermaur (2004) studied
perceptions of crime trends over the previous two years in New South Wales and
Western Australia and found that respondents have a proclivity to perceive crime is
rising. In a more recent New South Wales study, Jones, Weatherburn and MacFarlane
(2008: 6) found that only 11 percent correctly identified that property crime had
decreased in the five years prior to interview.
The third crime knowledge question asked was „What percentage of recorded crime
do you think involves violence or the threat of violence?‟ Respondents had the choice
of one quarter or less, between one quarter and a half, between a half and three
quarters and more than three quarters. The Crime and Sentencing booklet explained
that crimes of violence comprise only about 13 percent of crimes recorded by the
police. All traffic offences were excluded from this calculation, although drinkdriving offences could legitimately be regarded as criminal offences if not crimes.
Figure 5: Recorded crime in Tasmania: distribution of offences and perceptions
of crimes of violence
other offences 5%
fraud and similar 3%
other property 6%
mv stealing 5%
injury to property 15%
burglary 18%
stealing 35%
person 13%
Figure 6: Perceptions of proportion of crime that is violent
Don't know (8% jurors)
1/4 or less (17% jurors)
1/4 to 1/2 (34% jurors)
1/2-3/4 (30% jurors)
>3/4 (11% jurors)
Source: recorded crime data from Tasmania Police 2005-2006.
38
Part 3: Results
As can be seen in Figures 5 and 6, respondents thought a far larger proportion of
crime involved violence than is the case. Similar misperceptions have been found in
other studies (Gelb 2006: 13, 30). For example, in a New South Wales study, Jones et
al (2008: 6) found that more than 96 percent overestimated the proportion of crimes
that involve violence (they were asked to nominate the proportion of police recorded
crime that involved violence rather than being offered a range). In Indermaur‟s (1987)
Perth study, 73 percent of respondents substantially overestimated the proportion of
crimes involving violence. And AuSSA 2007 found that less than four percent of
survey respondents were accurate in their knowledge of the proportion of crime that
involves violence (Roberts and Indermaur 2009: 10).
Respondents were asked about their knowledge of sentencing practices, namely the
proportion of convicted offenders who were sent to prison for burglary and rape.
Table 12 indicates that 71 percent of respondents under-estimated the immediate
imprisonment rate for rape and 80 percent did so for burglary. Similar under-estimates
of imprisonment rates have been reported in other studies. In the 2008 New South
Wales study noted above, 89 percent of respondents nominated a figure lower than
the correct proportion (61%) (Jones et al 2008: 7).
Table 12: Perception of imprisonment rates for burglary and rape, percent
Proportion of convicted offenders sent to prison
Crime
Burglary
Rape
a
accurate response
0-25%
36
11
26-50%
44
28
51-75%
a
17
32
75% +
3
a
29
To get an indication of the degree of overestimation by jurors, the figure below shows
the actual penalty range created using the Supreme Court‟s sentencing database of
individual sentencing comments. The figure shows that the immediate imprisonment
rate for rape is more than 90 percent and for burglary it is more than 60 percent.1
1 Burglary is a crime that is triable summarily if the defendant so elects and the value of the property stolen is not
more than $20,000. So there are many cases of burglary that are heard in the Magistrates Court. However, data on
sentencing patterns for one incident of burglary in Magistrates Courts is available and shows that the rate of
immediate imprisonment is 50.1 percent (Tasmania Law Reform Institute 2008: 64). It follows that,
notwithstanding the fact that there are less burglaries heard in the Supreme Court, the immediate imprisonment
rate for burglary is 51-75 percent.
39
Jury Sentencing Survey
Figure 7: Sentences types by offence 2001-2006
What is the relationship between crime and sentencing knowledge and sentencing
patterns and sentencing preference?
Is there a relationship between those who think crime rates are rising and those who
chose a tougher sentence than the judge? Is there a relationship between those who
under-estimate sentencing severity and their choice of sentence? Previous research
has shown that public misperceptions are associated with a belief that sentences are
too soft. People who know more about crime rates are less punitive, (Roberts and
Indermaur 2007: 61-62) and this variable (thinking there is a lot more crime) was the
most powerful predictor of a belief that sentences were far too soft in the 1996 British
Crime Survey (BCS) (Hough and Roberts 1999: 18).
A bivariate analysis found that those jurors who thought recorded crime had increased
in the last five years were a little more likely to have selected a more severe sentence
than the judge compared with those who thought it had stayed the same or decreased,
however, the difference was not statistically significant. Table 13 shows that those
who accurately responded that one quarter or less of recorded crimes involved
violence were more likely to have chosen a more lenient sentence than the judge than
those who over-estimated incidence of violent crime. Therefore, on these measures of
knowledge (recorded crime rate trends, imprisonment rates and proportion of crime
that involves violence), the more knowledgeable participants were not significantly
less punitive than the less knowledgeable using Stage 1 comparative sentence severity
as the measure of punitiveness.
40
Part 3: Results
Table 13: Perception of crime that involves violence and severity of sentence
Assessment of proportion of crime that
involves violence
Pre information comparative severity (all offenders and
all jurors)
Less severe
Same
More severe
%
%
%
Accurate estimate (N = 127)
57.5
4.7
37.8
Over-estimate (N = 571)
50.4
3.5
46.1
Total (N = 698)
51.7
3.7
44.6
8. Perceptions of risk of victimisation and fear of crime
In Questionnaire 1 respondents were asked to estimate the risk of being a victim of
burglary, motor vehicle theft, assault and robbery in the next 12 months with interval
responses in five categories. The actual risk is depicted in the following table. The
risk of being burgled is likely to be two - five percent, the risk of motor vehicle theft
is lower and the risk of robbery is even lower still. For assault the risk is about five
percent or 1 in 20 but the risk decreases from the age of 20 and is lower for females.
Table 14: Crime victimisation rates for households and individuals
Tasmania
2002
%
Household victims
Break and enter
5.2
Motor vehicle theft
1.6
Personal victims
Assault
5.0
Robbery
0.3
Data Source: ABS, Crime and Safety Australia 2005.
As percentage of all households
Australia
2005
%
2002
%
2005
%
2.1
0.9
4.7
1.8
3.3
1.0
4.4
0.1
4.7
0.6
4.8
0.4
As Table 15 indicates, jurors tended to overestimate their risk of being a victim of
crime on all categories with nearly 20 percent of respondents thinking they had a 50
percent or higher risk of being burgled.
Table 15: Juror Perception of Risk of Victimisation
Level of Risk
Less than 6 percent
6-10 percent
11-30 percent
31-50 percent
>50 percent
Total
a
totals do not equal 100% due to rounding
Burglary
%
33
31
18
12
7
a
100
M/V theft
%
38
28
19
10
5
100
Assault
%
55
27
12
5
2
a
100
Robbery
%
41
33
16
7
3
100
A combined risk of victimisation variable was constructed for comparing whether
respondent‟s estimated risk of victimisation was associated with punitiveness as
measured by the Stage 1 comparative sentence severity variable. Bivariate analysis
showed no statistical link between this measure of punitiveness and perceptions of the
risk of victimisation.
41
Jury Sentencing Survey
Jurors were also asked about their feelings of personal safety in two questions: „How
safe do you feel walking alone in your area after dark?‟ (C2). And, „How safe do you
feel at home alone at night?‟ (C3). As the table below shows, respondents most
commonly felt „fairly safe‟ walking alone after dark and „very safe‟ at home alone at
night, with one third feeling unsafe walking alone at night.
Table 16: Perceptions of safety
Walking alone
after dark
Home alone at
night
Very safe
%
Fairly safe
%
A bit unsafe
%
Very unsafe
%
20
47
27
6
48
43
8
1
Sprott and Doob (1987) found that as fear increased, the proportion of people thinking
sentences were too lenient increased. Similarly, results from AuSSA 2007 showed a
statistically significant (although weak) relationship between fear of crime measures
and agreement with a need for stiffer sentences (Roberts and Indermaur 2009: 16). In
the present study, cross-tabulating levels of fear with punitiveness as measured by the
Stage 1 comparative sentence variable showed no significant differences. For
example, of those who felt very safe walking alone at night, 58 percent chose a less
severe sentence than the judge and 38 percent a more severe sentence. But those who
felt very unsafe selected very similar proportions of less severe (54%) and more
severe sentences (39%). Similarly, those who felt very safe at home alone at night
chose similar proportions of less severe (56%) and more severe (41%) sentences as
those who felt a bit unsafe (53% and 45%). Nor was a significant link found between
sentence choice and fear when the two questions were used to create a fear scale.
The relationship between fear and beliefs about crime trends and sentencing was also
examined. Those who felt safe walking at night were significantly more likely to
think that crime had decreased or stayed the same (p=.017). They were also more
likely to accurately estimate the proportion of reported crime that involves violence
(p=.005) and to answer the rape imprisonment rate correctly (p=.000). There was a
similar relationship between feelings of safety at home alone at night and crime and
sentencing knowledge. The relationship between fear and perceptions of risk of
victimisation was also examined. Those who were less fearful were more likely to
accurately estimate the risk of being a victim of home burglary, motor vehicle
stealing, assault or robbery.
9. Jurors’ general opinion of current sentencing practices
In the light of the limitations of answering a single question about the severity of
sentencing levels (Roberts and Stalans 1997), the study asked: „In general would you
say that current sentences for the following crime types are much too tough, a little
too tough, about right, a little too lenient or much too lenient?‟ Table 17 shows that,
across all offence types, the majority responded that sentences were too lenient. This
was most pronounced for sex and violent offences with 80 percent and 76 percent of
responses saying sentences were too lenient. Few respondents thought sentences were
too tough, although drug offences attracted a higher „too tough‟ response than the
other offences. Jurors were most satisfied with sentences for property offences. This
suggests that the public is more divided about drug offences than the other categories,
with a more even split between about right, a little too lenient and much too lenient
42
Part 3: Results
responses. These results are consistent with other research (see Doob and Roberts
1983; Indermaur 1987) which found that respondents who felt that sentences were too
lenient, tended to have in mind the worst kind of offenders such as sex or violent
offenders.
Table 17: Are current sentencing practices too tough/lenient?
Opinion on
Type of Crime
Sentence
a
N = 674-681
Sex
Violence
Drugs
Much too tough
1
0
2
A little too tough
1
1
6
About right
18
23
36
A little too lenient
39
53
35
Much too lenient
41
23
21
Total
100
100
100
a
N = 674 (sex), 681 (violence), 677 (drugs) and 674(property).
Property
0
4
42
39
15
100
Do jurors’ opinions differ depending on the crime type of their trial?
Juror respondents differ from respondents in representative surveys in that they are
being asked about sentencing leniency in the context of a real trial. The type of crime
that they tried may influence their response to the general question about sentencing
severity. In other words, what is the response of sex offence jurors, for example, to
the general question about sentencing sex offences? Does it differ from the responses
of the other jurors? Being involved in a trial of a sex offender may have an impact on
the juror‟s perception of a stereotypical sex offender and affect the answer to the
abstract question about sentencing severity for sex offenders.
Table 18 shows that sex offence trial jurors were a little less likely to say sentences
for sex offences were too lenient compared with other jurors. The same is true for
drug and property offences but not for violent offences. The interpretation of this
finding will be further explored below in Part B of this Part and in Part 4.
Table 18: General perceptions of sentencing leniency by respondent’s trial type
Juror’s trial
type
Sex offence
Violent
offence
Drug offence
Property
Jurors
Sex offence trial jurors
Other jurors
All jurors
Violent trial jurors
Other jurors
All jurors
Drug trial jurors
Other jurors
All jurors
Property trial jurors
Other jurors
All jurors
Too tough
%
2.4
1.3
1.5
0
1.3
0.9
7.9
7.8
7.8
0.9
4.7
4.2
About right
%
23.6
17.2
18.4
22
23.9
23.2
41.4
34.1
35.6
49.5
40.6
42.0
Too lenient
%
73.9
81.5
80.1
78
74.7
75.9
50.7
58.1
56.6
49.5
54.7
53.9
Table 19 looks at the relationship between general attitudes to sentencing severity and
the Stage 1 comparative sentence severity variable. It is reasonable to assume that the
two are correlated; that those who chose a more severe sentence than the judge are
more likely to have said sentences are too lenient than those who chose a less severe
sentence. The results of this analysis confirm that this is indeed the case. Those who
chose a less severe sentence than the judge were less likely to say sentences for each
of the categories of offences were too lenient.
43
Jury Sentencing Survey
Table 19: Comparative severity of juror’s sentence by view on current
sentencing patterns
Too
tough
%
Severity
of juror's
sentence
compared
to Judge
About
right
%
Too
lenient
%
Total
N
Q1: Are current sentences for violence offences too tough/lenient? (all offenders)
Less severe
0.6
26
73.5
(358)
0
26.9
73.1
(26)
More severe
1.3
18.2
80.5
(308)
Totals
0.9
22.5
76.6
(692)
Same
Q1: Are current property sentences too tough/lenient? (all offenders)
Severity
of juror's
sentence
compared
to Judge
Less severe
5.1
41.9
53
(353)
0
53.8
46.2
(26)
More severe
3.3
40.7
56.1
(305)
Totals
4.1
41.8
54.1
(684)
Same
Q1: Are current drug sentences too tough/lenient? (all offenders)
Severity
of juror's
sentence
compared
to Judge
Less severe
11.2
37.4
51.4
(356)
Same
0.0
48.0
52.0
(25)
More severe
5.6
32.4
62.1
(306)
Totals
8.3
35.5
56.2
(687)
Q1: Are current sex sentences too tough/lenient? (all offenders)
Severity
of juror's
sentence
compared
to Judge
Less severe
1.7
19.3
79.0
(353)
Same
0.0
26.9
73.1
(26)
More severe
1.6
16.7
81.6
(305)
Totals
1.6
18.4
80.0
(684)
A punitiveness scale was created using jurors‟ responses as to whether sentences were
too tough, about right or too lenient across the four offence categories (the Stage 1
general punitiveness index). A comparison of mean scores on this index with the
Stage 1 comparative sentence severity variable showed that those who wanted a more
severe sentence than the judge had statistically higher mean scores on the general
punitiveness index than those who wanted a less severe sentence than the judge
(p=.006). In other words they were more likely to consider current sentencing too
lenient. Mean general punitiveness index scores were also compared with levels of
knowledge about crime, sentencing, perceptions of risk and fear. This showed that:
44

those who thought crime had decreased had a lower punitive score than those
who thought it had increased (p=.000);

those who over-estimated the percentage of crime that involves violence had
a higher score on the general punitive index than those who correctly
estimated it (p=.000);

those who correctly estimated the proportion of convicted rape offenders
who were sentenced to prison had a lower score than those who underestimated it (p=.018), but there was no significant correlation between
estimates of burglary imprisonment rates and mean scores on the general
punitive index;
Part 3: Results

those who were more accurate in their estimation of the risk of victimisation
had lower punitiveness scores (p=.000); and

those who were more fearful were more punitive than those who were less
fearful (p=.000).
These findings are consistent with previous research which showed that public
misperceptions were associated with a belief that sentences were too lenient.
Misperceptions associated with a belief that sentences were too lenient in Britain
included a belief that there was a lot more crime; over-estimating the proportion of
crime that involves violence; and under-estimating the proportion of convicted
muggers and burglars who go to prison (Hough and Roberts 1999: 18). In Australia,
results from AuSSA 2003 (Roberts and Indermaur 2007) and AuSSA 2007 showed
similar findings. For example, AuSSA 2007 showed the desire for stiffer penalties
was significantly positively correlated with beliefs that crime was increasing and with
the number of reported crimes that involved violence, and negatively correlated with
the proportion of men convicted for assault and home burglary who were imprisoned
(Roberts and Indermaur 2009: 19).
10. Jurors’ opinions of judges before sentence (Stage 1)
A common question in representative surveys such as the BCS asks respondents how
„in touch‟ judges are with what ordinary people think (Hough and Roberts 1998;
Mirrlees-Black 2001). This question is asked in the context of questions about the
criminal justice system and sentencing. Our question asked, „How in touch do you
think judges are with public opinion on sentencing?‟ Figure 8 below shows the
response given by the jurors to this question (B6) in Questionnaire 1. It shows that the
majority, 57 percent, responded that judges are „somewhat in touch‟ and that more
than two thirds of respondents thought judges were „in touch‟ (either very or
somewhat). Only 30 percent responded that they were „out of touch‟. This contrasts
with findings from two studies of the BCS which found a high proportion of
respondents, 46 percent, thought judges were very out of touch (Hough and Roberts
1998: 3) and 41 percent (Mirrlees-Black 2001: 5). Only 18 (Hough and Roberts 1998:
3) and 20 percent (Mirrlees-Black 2001: 5) reported that judges were in touch.
Perceptions of judicial remoteness have also been reported in Scotland (75 percent of
respondents thought judges were out of touch with what ordinary people think,
(Hutton 2005: 247)) and South Australia, with 73 percent of respondents in that State
agreeing with the statement „it is about time judges caught up with the real world
(Square Holes 2006). It is unsurprising, but nevertheless reassuring, that jurors have a
more favourable view of judges than the general public (see Benesh 2006).
45
Jury Sentencing Survey
Figure 8: How in touch are judges with public opinion in relation to sentencing
60.0%
Percentage
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Series1
Very in touch
n=87
Somewhat in
touch n=392
Somewhat out
of touch n=193
Very out of
touch n=16
12.6%
57.0%
28.1%
2.3%
Juror‟s perception of how in touch judges were with public opinion on sentencing was
then cross-tabulated with the juror‟s comparative sentence choice. Hough and Roberts
(1990: 15) have found that views regarding sentencers as being out of touch were
associated with a belief that sentences were too lenient. Thus, of the minority who
believed judges were in touch with society, 57 percent also felt that sentences were
too lenient. However, 90 percent of the respondents who thought that judges were out
of touch also felt that sentences were too lenient. As shown in Table 20, punitiveness
measured by sentence choice (Stage 1 comparative sentence severity) showed a much
weaker relationship with a belief that judges were out of touch in our study.
Table 20: Jurors view of in/out of touch and comparative sentence choice
Severity of proposed sentence compared to judge imposed sentence
Less severe
%
Same
%
More severe
%
Very in touch all offenders
55.7
4.5
39.8
Somewhat in touch all offenders
51.6
4.0
44.3
Somewhat out of touch all offenders
50.6
2.8
46.6
Very out of touch all offenders
44.1
2.9
53.0
Total
%
(N)
100
(88)
100
(397)
100
(178)
100
(34)
But what of the relationship between views that judges are out of touch and a belief
that sentences are too lenient? Table 21 shows that our study produces similar results
in this comparison to those of Hough and Roberts (1990). Of the minority who
thought judges were very in touch with public opinion on sentencing, two thirds said
sentences for sex offences were too lenient compared with 94 percent of those who
said judges were very out of touch. Similar patterns were found for other crime types.
46
Part 3: Results
Table 21: Relationship between judicial remoteness and perceptions that
sentences are too lenient (Stage 1)
Remoteness of Judges from Public Perceptions on
Sentencing
Sex
%
Violence
%
Property
%
Drugs
%
Very in touch
66.3
57.6
38.8
40.0
Somewhat in touch
78.9
72.6
49.9
52.2
Somewhat out of touch
87.4
89.6
66.7
71.4
Out of touch
93.8
87.5
75.0
68.8
Total (too lenient)
80.1
75.9
53.8
56.5
Note: The percentages in this table are percent of total in/out of touch opinion for that offence.
B. STAGE 2
1. Who were the Stage 2 respondents?
Nearly two thirds, or 64 percent, (N = 445 jurors), from Stage 1 participated in Stage
2 by returning Questionnaire 2. Of these, 212 also offered to be part of the
interviewing phase of the study, and 50 of these jurors were subsequently
interviewed. As these jurors were from 104 of the 138 trials in which at least one
Stage 1 response was received, the question of whether the response rate for offence
type would differ between Questionnaire 1 and 2 arose. As shown in Table 22 the
Stage 2 response rate was much the same for sex, violence, drug and property
offences although the rate for violent offences was a little stronger. The response rate
was the poorest for culpable driving but with just three culpable driving trials in the
study, the numbers were small for this offence category.
Table 22: Participation by type of offence, Stage 1 and Stage 2 compared
Type of offence
Sex
Violence
Drugs
Property
Culpable Driving
Other
Total
Number of Q1
responses
128
253
143
113
8
53
698
Q1 response rate %
35.6
36.9
33.1
37.7
22.2
40.2
35.9
91
153
94
69
6
32
445
20.4
34.4
21.1
15.5
1.3
7.2
100
Number of Q2
responses
% of responses
The socio-demographic profile of the two groups was also compared. As shown in the
Table below, female jurors were more likely to complete Questionnaire 2 than males,
increasing the over-representation of females in Stage 2. The same tendency for ages
18-24 and 25-44 to be under-represented also increased in Stage 2, as did the
tendency for the age group 45-64 to be over-represented. However, the underrepresentation of the 65 + age group decreased slightly. In summary, female jurors
and those aged over 45 seemed to be more willing to participate in Stage 2 as did
those with more education.
47
Jury Sentencing Survey
Table 23 Comparison on jurors who completed only Questionnaire 1 with those
that also completed Questionnaire 2
Respondents who completed
Questionnaire 1 only
%
n
52
(130)
48
(120)
100
(250)
Respondents completed
Questionnaire 1 & 2
%
n
43.6
(194)
56.4
(251)
100
(445)
Gender
Male
Female
Total
Age
18 – 24
25 – 44
45 – 64
65+
Total
17
37.7
40.5
4.9
100
(42)
(93)
(100)
(12)
(247)
6.3
30.7
52.6
10.4
100
(28)
(136)
(233)
(46)
(443)
Year 10 or below
Year 11
Year 12
Trade
Certificate or Diploma
27.3
6.8
16.1
10.8
22.1
16.9
100
(68)
(17)
(40)
(27)
(55)
(42)
249
18.7
3.6
10.6
11.9
26.4
28.8
100
(83)
(16)
(47)
(53)
(117)
(128)
444
Education
Bachelor's degree or above
Total
Punitiveness of Stage 2 Respondents
A comparison of Stage 1 and Stage 2 respondents on comparative sentence severity
found an almost identical pattern of responses. This suggests that on this measure
(Stage 1 comparative sentence severity) there was minimal difference between Stage
1 respondents and Stage 2 respondents. Therefore, results from Questionnaire 2 are
unlikely to be skewed significantly by the lower sample numbers from Questionnaire
1.
Comparing the judge‟s sentence with Stage 1 juror‟s sentence choice by type of
crime, showed (see Figure 3) that less than half of jurors were more severe than the
judge for sex, violence and property offences and just 50 percent more severe than the
judge for drug offences. Figure 9 looks at Stage 2 respondents only and indicates
Stage 2 jurors were:
48

less severe than the judge for most offence categories but the exceptions are
violence and „other‟ rather than drugs and other in Figure 3;

more lenient than the judge for property and culpable driving offences to the
same degree as Stage 1 respondents; and

least likely to be less severe than the judge for „other‟ offences and a little
less likely than Stage 1 jurors to be less severe for violent offences.
Part 3: Results
Percentage
Figure 9: Severity of juror’s proposed sentence compared with judge’s sentence
(percent) by type of offence
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Less severe
All
offences
N = 454
Sex
offences
n=84
Violent
offences
n=168
Drug
offences
n=93
Property
offences
n=68
Culpable
driving
n=8
Other
offences
n=33
51%
50%
45%
50%
67%
88%
42%
Same
4%
5%
6%
2%
2%
0%
6%
More severe
45%
45%
49%
48%
31%
12%
52%
2. Jurors’ views of the judges’ sentences
The first question in Questionnaire 2 asked, „Now that you know the sentence the
judge gave, was it what you expected?‟ with answer categories of „exactly‟, „close to
expected‟, „a little different‟, or „completely different‟. For two thirds of Stage 2
jurors, the actual sentence was about what they expected (exactly as expected 11
percent and close to expected 57 percent). Around a quarter reported that they
expected the sentence to be „a little different‟ and only seven percent thought the
sentence would have been completely different. Approximating our responses to fit
those from the UK Crown Court Study (Zander and Henderson 1993: 223), similar
proportions stated that the sentence was expected.
Given the differences between the crime categories in the comparison between judge
and juror sentences (see Figure 3), it might be expected that Stage 2 respondents
would also be more likely to find property offence sentences different than expected
when compared with sex, violent and drug sentences. Figure 11 shows that this was
not the case. There was little difference in the major categories of sex, violence, drugs
and property in expectations.
49
Jury Sentencing Survey
Percentage
Figure 10: Was the sentence expected by jurors? (by offence type)
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Sex n=90
Violence
n=153
Drugs
n=94
Property
n=68
Culpable
driving
n=6
Other
n=32
Exactly as Expected
11.1%
13.7%
10.6%
7.4%
16.7%
6.3%
Close to Expected
55.6%
54.2%
54.3%
60.3%
83.3%
65.6%
A little different
24.4%
24.2%
29.8%
25.0%
.0%
25.0%
Completely different
8.9%
7.8%
5.3%
7.4%
.0%
3.1%
Jurors were next asked about their views as to the appropriateness of the sentence.
Almost 90 percent of Stage 2 respondents rated the judge‟s sentence as appropriate,
with responses evenly split between very appropriate and fairly appropriate. Almost
half were „very satisfied‟ with the sentence and only a very small percentage (1.6%)
thought it „very inappropriate‟. This result is interesting in that it indicates that overall
satisfaction levels are high even though almost half of jurors had selected a more
severe sentence than the judge.
Comparison of satisfaction with sentence across offence type indicated some
differences in attitude. The data summarised in Figure 11 suggests that respondents
were less satisfied with drug and sex offence sentences than those for property and
violence. A smaller proportion of „very appropriate‟ responses were made for drugs
and sex offences and there were smaller proportions of „very appropriate‟ and „fairly
appropriate‟ sentences combined. However, the differences were not significant
(p=.057). In addition, there were more „inappropriate‟ responses for drug offences.
Percentage
Figure 11: How appropriate was the sentence for each crime type
50
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
.0%
Sex N=90
Violence
N=151
Drugs
N=94
Property
N=69
Culpable
driving
N=6
Other
N=32
Very appropriate
35.6%
50.3%
35.1%
56.5%
66.7%
46.9%
Fairly appropriate
52.2%
41.7%
47.9%
36.2%
33.3%
46.9%
Fairly inappropriate
8.9%
6.0%
17.0%
5.8%
.0%
6.3%
Very inappropriate
3.3%
2.0%
.0%
1.4%
.0%
.0%
Part 3: Results
The data were further analysed to see if the greater dissatisfaction with drug offences
was due to the fact that respondents were more divided about whether they were too
tough and too lenient. However, analysing the sentence choice of the „inappropriate‟
responses suggests that the majority (11/14) of „fairly inappropriate‟ responses to drug
offence sentences was because they were too lenient with most respondents having
selected a more severe sentence than the judge.
The next question that arose was whether satisfaction level varied between those who
chose a more lenient sentence than the judge, and those who chose a more severe
sentence. Analysis shows (see Figure 12) that those who selected a more lenient
sentence were more likely to say the judge‟s sentence was „very appropriate‟ than
those selecting a tougher sentence. They were also less likely to say the sentence was
inappropriate. This suggests that jurors are more tolerant of differences between the
judge‟s sentence and the juror‟s sentence if the judge was tougher than they were but
less tolerant of a sentence that was more lenient. Of course jurors were most satisfied
if the judge‟s sentence was the same2.
Figure 12: Satisfaction levels and sentence choice (all offences)
70.0%
60.0%
Percentage
50.0%
40.0%
30.0%
20.0%
10.0%
.0%
Less severe
N=209
Same N=18
More severe
N=181
Very appropriate
53.1%
61.1%
34.8%
Fairly appropriate
42.6%
38.9%
46.4%
Fairly inappropriate
3.3%
.0%
16.0%
Very inappropriate
1.0%
.0%
2.8%
The next table looks at levels of satisfaction with the judge‟s sentence and Stage 1
comparative sentence severity (or sentence choice) for the four main offence
categories. As Table 24 shows, for each offence type, those selecting a less severe
sentence were more likely to say the sentence was „very appropriate‟ compared with
those selecting a more severe sentence. This is particularly so for sex offences where
nearly half of those who had selected a more lenient sentence at Stage 1 said that the
judge‟s sentence was „very appropriate‟ compared to 19 percent of those who had
selected a more severe sentence. It was least likely for property offences. Moreover,
those who selected a more severe sentence were more likely to say that the sentence
was fairly or very inappropriate compared with those who selected a more lenient
sentence. For example, in drug offence cases 27 percent of those selecting a more
severe sentence said the sentence was fairly inappropriate compared to seven percent
who had selected a more lenient sentence.
2 Before exploring these findings for different offences it should be noted that a small proportion of respondents (6/408) had
selected the same sentence and said that it was fairly appropriate. Five of these at Stage 2 had changed their mind and now would
have preferred a more severe sentence and one comment related to the verdict rather than the sentence.
51
Jury Sentencing Survey
Table 24: Juror’s response to judge’s sentence by sentence choice
Offence
type
N = 378
Sex
n=83
Violence
n=140
Drugs
n =93
Property
n=62
a
Sentence
choice
Less severe
More severe
Same
sentence
Less severe
More severe
Same
sentence
Less severe
More severe
Same
sentence
Less severe
More severe
Same
sentence
Very
appropriate
%
48
19
Fairly
appropriate
%
48
57
Fairly
inappropriate
%
2
19
Very
inappropriate
%
2
5
25
75
0
0
100
59
44
33
42
5
10
3
4
100
100
60
40
0
0
100
41
32
52
42
7
27
0
0
100
a
100
50
17
33
0
100
54
43
46
38
0
14
0
5
100
a
100
100
0
0
0
100
Total
100
100
Due to rounding some totals do not equal 100%.
As a follow-on from the question about the appropriateness of sentence, jurors were
asked, unless they thought the sentence was very appropriate, to indicate what they
thought the sentence should have been (A3). A variable was constructed from the
responses comparing the severity of judge‟s sentence with the juror‟s view. This
variable is called „Stage 2 comparative sentence severity‟. Within this variable, those
who said the sentence was very appropriate or added a supportive comment were
entered as the „same‟ as the judge. Those who indicated that a longer prison sentence,
larger fine or a more severe sentencing option was appropriate were entered as „more
severe‟. If an additional sentencing option was indicated then this was classified as
„more severe‟. And those who indicated a shorter prison sentence, more suspension
than the judge or a more lenient sentencing outcome were entered as „less severe‟.
The distribution of the categories on the Stage 2 comparative sentence severity
variable are „less severe‟ = 9 percent; „same‟ = 53.5 percent; and „more severe‟ = 37.5
percent.
Cross-tabulating this variable by type of offence shows that in Stage 2:

a majority were satisfied with the same sentence as the judge for violence
and property offences;

less than 10 percent of respondents wanted a less severe sentence than the
judge across the four major offence categories;

a significant proportion of respondents wanted a more severe sentence for
sex offences and drug offences; and

respondents were least likely to want a more severe sentence for property
offences.
This variable confirms the results from Figure 11 that jurors are least satisfied with
sex and drug offence sentences and are most satisfied with property offence
sentences.
52
Part 3: Results
Percentage
Figure 13: Stage 2 comparative sentence severity by offence type
90.0%
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
.0%
Sex
n = 91
Violence
n = 153
Drugs
n = 94
Property
n = 67
Culpable
driving
n=6
Other
n = 32
Less severe
9.9%
8.5%
8.5%
9.0%
16.7%
9.4%
Same
44.0%
56.2%
45.7%
62.7%
83.3%
65.6%
More severe
46.2%
35.3%
45.7%
28.4%
0%
25.0%
Figure 12, detailed earlier, found a greater acceptance of more lenient sentences than
more severe sentences using the responses about appropriateness of the sentence.
Table 25 confirms this finding. Using the Stage 2 comparative sentence variable as
the independent variable, it was found that those whose sentence choice was more
lenient in Stage 1 were more likely to agree with the judge‟s sentence compared with
those who were more severe. Specifically, 57 percent of those who selected a more
severe sentence still wanted a more severe sentence after knowledge of the judge‟s
sentence and receipt of the booklet compared with 18 percent of those who still
wanted a more lenient sentence. This suggests that those who are more punitive have
less malleable views than those who are more lenient. Interestingly, of those who had
chosen a less severe sentence than the judge, one fifth subsequently decided they
wanted a more severe sentence than the judge. The fact that some of those who had
selected the same sentence as the judge subsequently changed their minds and wanted
a more severe sentence has been adverted to above in the discussion of Figure 12.
That jurors who selected a more lenient sentence than the judge in Stage 1 are more
likely to agree with the judge‟s sentence than those selecting a more severe sentence
also means that on this measure (Stage 2 comparative sentence severity) some
respondents became more punitive. In total 43 percent of respondents have become
more punitive in Stage 2 than they were at Stage 1 with 19 percent becoming less
punitive than they were at Stage 1.
53
Jury Sentencing Survey
Table 25: Stage 1 comparative sentence severity compared with Stage 2 juror
response to judge’s sentence
Offender 1 data only
p =.000
Stage 1:
comparative
sentence
severity
Less severe
%
Stage 2 response to judge’s sentence
Same
More severe
%
%
Total
%
Less
severe
17.3
(36)
60.6
(126)
22.1
(46)
100.0
(208)
Same
0.0
(0)
72.2
(13)
27.8
(5)
100.0
(18)
More
severe
2.2
(4)
40.4
(74)
57.4
(105)
100.0
(183)
Total
9.8
(40)
52.1
(213)
38.1
(156)
100.0
(409)
Stage 2 response to judge’s sentence
All offenders
Stage 1:
comparative
sentence
severity
Less severe
%
Same
%
More severe
%
Total
%
Less
severe
18.0
(41)
59.6
(136)
22.4
(51)
100
(228)
Same
0.0
(0)
73.7
(14)
26.3
(5)
100
(19)
More
severe
2.4
(5)
40.2
(82)
57.4
(117)
100
(204)
Total
10.2
(46)
51.4
(232)
38.3
(173)
100
(451)
Judges’ sentencing remarks
Questionnaire 2 asked respondents about the judge‟s sentencing comments. In answer
to „Was there anything in the judge‟s sentencing remarks that you particularly
disagree with?‟, only 17 percent of Questionnaire 2 jurors (N = 445) indicated
disapproval of the judge‟s sentencing comments. Many of these comments did not
relate to sentence, e.g. „We as jurors were not made aware of the convicted man‟s
prior convictions and outcomes‟. The most common criticism disagreed with the
judges interpretation of the facts, another attacked the severity of the sentence
imposed, and others disagreed about whether a factor should have been aggravating or
mitigating, for example, one suggested the offender‟s record should have been
aggravating and another that more weight should have been given to the offender‟s
vulnerability. The same respondent thought the judge should have shown more
compassion by his tone of voice. A couple of respondents disagreed with the purpose
of sentencing expressed by the judge, two suggested the judge should have
condemned the offence in stronger terms and one wanted an alternative sentencing
option.
In answer to Questionnaire 2 A5, „Was there anything in the judge‟s sentencing
remarks that you particularly agree with?‟ about a third either expressly stated that
there was nothing they specifically agreed with or this was implied by their failure to
complete the question. Another third particularly agreed with the judge‟s summary of
the facts (including appraisal of the accused), some stated general approval of the
comments, others mentioned the appropriateness of the judge‟s sentence, and a few
expressed agreement with the sentencing goals referred to by the judge.
54
Part 3: Results
3. Sentencing Goals
In Questionnaire 2 jurors were asked to rank sentencing goals from a list of seven
commonly recognised goals from the most important (1) to the least important as they
thought they pertained to their case (7). As shown in Table 26 punishing the offender
was the most important sentencing goal, followed by specific deterrence.
Denunciation (expressing community disapproval), general deterrence and
rehabilitation were each regarded as most important by about 10 percent of
respondents. Specific deterrence was most often ranked second, general deterrence
third, followed by rehabilitation and denunciation fourth. Not surprisingly our jurors
ranked as the most important goals: punishment, specific deterrence and general
deterrence.
Table 26: Ranking of sentencing goals
Rank
1
Rank
2
Rank
3
Rank
4
Rank
5
Rank
6
Rank
7
Punishing the offender
40.6
13.8
9.8
10.4
9.0
10.7
6.2
Separating the offender from society
6.0
12.4
6.3
10.6
16.1
19.7
28.7
Expressing community disapproval
10.6
12.9
17.1
17.6
16.4
13.5
11.6
Assisting the offender's rehabilitation
8.3
Deterring others from committing
9.2
similar crimes
Deterring the offender from re-offending
21.8
Compensating the victim(s) or the
3.7
community
Darker cells indicate highest ranking of item.
13.1
14.1
17.6
16.4
16.9
13.7
18.0
23.4
16.7
13.5
14.5
4.7
25.1
19.4
15.1
11.6
4.8
2.4
4.7
9.8
12.0
17.1
20.0
32.7
The next series of diagrams look at the most important sentencing goal for the offence
categories of sex, violence, drugs and property. They show that the pattern of
responses was similar for sex and violent offences, with punishment and specific
deterrence most often selected as the most important sentencing goal. For drug
offences, specific deterrence was the most commonly selected sentencing goal
followed by punishment and general deterrence and for property offences, it was
punishment followed by denunciation and rehabilitation.
Figure 14: Most important sentencing goal for sex offenders
60.0%
50.0%
Percentage
40.0%
30.0%
20.0%
10.0%
.0%
Sex N91
Punishing the
offender
Separating the
offender from
society
Expressing
community
disapproval
Assisting the
offender's
rehabilitation
Deterring
others from
committing
similar crimes
Deterring the
offender from
re-offending
Compensating
the victim(s) or
the community
49.5%
7.7%
11.0%
3.3%
11.0%
14.3%
3.3%
55
Jury Sentencing Survey
Percentage
Figure 15: Most important sentencing goal for violent offenders
45.0%
40.0%
35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
.0%
Punishing the
offender
Separating the
offender from
society
Expressing
community
disapproval
Assisting the
offender's
rehabilitation
Deterring
others from
committing
similar crimes
Deterring the
offender from
re-offending
Compensating
the victim(s) or
the community
41.4%
9.9%
13.2%
5.9%
5.3%
21.7%
2.6%
Violence N152
Percentage
Figure 16: Most important sentencing goal for drug offenders
45.0%
40.0%
35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
.0%
Drug N90
Punishing the
offender
Separating the
offender from
society
Expressing
community
disapproval
Assisting the
offender's
rehabilitation
Deterring
others from
committing
similar crimes
Deterring the
offender from
re-offending
Compensating
the victim(s) or
the community
26.7%
3.3%
3.3%
12.2%
14.4%
38.9%
1.1%
Percentage
Figure 17: Most important sentencing goal for property offenders
40.0%
35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
.0%
Property N67
56
Punishing the
offender
Separating the
offender from
society
Expressing
community
disapproval
Assisting the
offender's
rehabilitation
Deterring
others from
committing
similar crimes
Deterring the
offender from
re-offending
Compensating
the victim(s) or
the community
37.3%
1.5%
19.4%
13.4%
7.5%
10.4%
10.4%
Part 3: Results
4. Aggravating and Mitigating Factors
In Questionnaire 2 (A7) the most commonly recognised aggravating and mitigating
factors were listed and respondents were asked to go though the lists and indicate
whether the factor was very important, quite important, not very important,
unimportant or did not arise. The table below lists the aggravating factors in order of
those most commonly occurring and shows the degree of importance or weight of
each factor by indicating the percentage of respondents who rated it as very
important, quite important and so on.
Table 27: Aggravating Factors, relative importance, percent
Juror
indicated a
factor
%
Very
Important
%
Quite
Important
%
Not very
Important
%
Unimportant
%
k. The injury, emotional harm, loss or
damage caused by the offence was
substantial
76.2
46.8
34.5
11.4
7.2
m. The offender had prior offences
65.5
44.2
35.1
11.6
9.1
a. The offence involved the use of actual
or threatened violence
50.9
70.0
22.0
4.9
3.1
g. The offender abused a position of trust
or authority in relation to the victim
49.8
58.0
25.6
8.2
8.2
l. The offence was part of a planned or
organised criminal activity
46.8
44.3
36.9
10.8
7.9
o. Other factors a juror considered
aggravated the seriousness of the
offence
46.0
45.5
31.5
7.7
15.4
i. The offence(s) involved multiple victims
or a series of criminal acts
45.7
39.8
42.8
11.9
5.5
f. The victim was vulnerable because very
old, very young, or because of a
disability
45.1
46.7
20.8
16.8
15.7
e. The offence was committed without
regard for public safety
42.2
34.1
34.1
18.9
13.0
c. The offence involved more than needed
violence
40.6
56.8
26.1
11.9
5.1
j. The offence was committed in the
company of other offenders
39.3
34.7
38.7
15.0
11.6
b. The offence involved use of a weapon
31.9
76.4
16.4
5.0
2.1
d. The offence was motivated by hatred or
prejudice against a member of a group
to which the offender believed the
victim belonged
20.0
29.5
28.4
22.7
19.3
n. The offender was on parole, subject to
a suspended sentence or on bail
18.4
49.4
33.7
9
7.9
h. The victim was a police officer or a
person exercising other public or
community functions and the offence
arose because of the victim’s
occupation
11.4
40.0
34.0
12.0
14.0
57
Jury Sentencing Survey
The four most frequently mentioned aggravating factors were that (k) the injury,
harm, loss or damage caused by the offence was substantial; (m) the offender had
prior convictions; (a) the offence involved actual or threatened violence, and (g) the
offender abused a position of trust or authority in relation to the victim. The factors
given the greatest weight when they did arise were (b) the offence involved use of a
weapon; (a) the offence involved actual or threatened violence; (g) there was abuse of
trust or authority; and (c) the offence involved gratuitous use of violence. Analysing
these four factors by type of offence showed that injury, harm, loss or damage and
breach of trust were most important in cases of sex offences. Prior convictions were
most important in cases of violent offences.
Table 28: Mitigating Factors, relative importance, percent
Juror
indicated a
factor %
Very
Important
%
Quite
Important
%
Not very
Important
%
Unimportant
%
h. The offender has good prospects of
rehabilitation either because of age or
otherwise
85.4
20.2
48.4
18.3
13
g. The offender is unlikely to reoffend
80.4
24.4
37.2
19.2
19.2
e. The offender was a person of good
character
74.7
14.9
33.1
27.5
24.5
d. The offender does not have any record
(or any significant record) of prior
convictions
65.6
27
34.9
17.3
20.8
a. The injury, loss or damage was not
substantial
65.5
15.7
29.5
24.9
29.9
k. The offender provided assistance to
law enforcement authorities
59.6
23.3
36.6
19.5
20.5
b. The offence was not part of a planned
or organised criminal activity
58.6
18.8
31.8
25.1
24.3
i. The offender has shown remorse for the
offence by making reparation for any
injury, loss or damage or in any other
manner
52.6
28.5
34
22.7
14.8
f. The offender was young (under 21) or
old (over 65)
47.8
12.9
25.0
28.4
33.6
47
19.2
28.8
19.7
32.3
39.4
16.3
30.2
23.3
30.2
j. The offender was not fully aware of the
consequences of their actions because
of the offender's age, mental disorder
or other disability
c. The offender was provoked by the
victim
The five most frequently mentioned mitigating factors were (h) good prospects of
rehabilitation; (g) the offender is unlikely to re-offend; and (d) the offender has no, or
no significant prior record and (a) the injury loss or damage was not substantial. The
highest rating factors for „very important‟ were (i) remorse; (d) no criminal record; (g)
the unlikelihood of re-offending; and (k) assistance to law enforcement authorities.
Good prospects of employment were more likely to be most important for property
and drug offences than for violence and sex offences and good character was least
important for sex offences.
58
Part 3: Results
To assist in answering the sixth research question (the extent to which the views of
jurors in the sentencing comments coincided or differed from those of the judge) it
was hoped to compare the jurors‟ responses in relation to the importance of
aggravating and mitigating factors with the views of the judge as indicated in the
Comments on Passing Sentence. However, an analysis of the sentencing comments
showed that while it was possible to see which factors the judge had mentioned, it
was not easy to determine the weight given to them by the judge. To determine this
would require that the judge, as well as the jurors, respond to the same question about
aggravating and mitigating factors and this was not done in this study.
5. The Booklet
The booklet „Crime and Sentencing‟ was sent out to jurors, who agreed to participate
in Stage 2, with Questionnaire 2 and the judge‟s sentencing comments. A copy of the
nine-page booklet is attached in Appendix 2. It has sections on measuring crime,
trends in recorded crime levels, the proportion of crime that involves violence and the
risk of victimisation. The sentencing section of the booklet includes a discussion of
the purposes of sentencing, relevant sentencing factors, the range of sentencing
options and some information on current sentencing practice. This sentencing
information was supplemented by a data sheet setting out the range of penalties
imposed for the particular crime for which the juror‟s defendant had been found
guilty.
A series of questions addressed the response of participating jurors to the booklet.
They were first asked (B1) about how well they had read the booklet. More than 60
percent read it in full and just two percent did not read it.
Figure 18: How well was the booklet read?
70.0%
Percentage
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Series1
Read the booklet
in full n=275
Read some
sections in full
n=103
Flick through the
booklet n=55
Not read the
booklet n=8
62.4%
23.4%
12.5%
1.8%
As a measure of how informative and useful the booklet was, participants were asked
(B2) what they planned to do with the booklet and as shown in Figure 19, most
planned to keep it.
59
Jury Sentencing Survey
Figure 19: Juror plans for booklet
80.0%
Percentage
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Series1
Keep the booklet
n=327
Throw it away
n=55
Give it to a friend
n=36
Other n=24
74.0%
12.4%
8.1%
5.4%
The booklet was designed to be easy to understand and accessible with an attractive
layout, and attempted to provide a factual and unbiased account of the material
covered. Participants (in Question B3) were asked to rate the booklet across a number
of dimensions from 1 (strongly agree) to 5 (strongly disagree). The mean rating on
each of these dimensions is shown in Figure 20. On five of the six scales respondents
rated the booklet higher than two. Jurors also indicated that the booklet made them
feel more confident in the criminal justice system.
Figure 20: Mean ratings for the booklet on various dimensions
Participating jurors were asked how interested jurors (in general) would be in
receiving the booklet after participating in a trial which resulted in a guilty verdict.
Most thought jurors would be very/quite interested (97%). To the next question on
whether jurors would be interested in knowing how to access the judge‟s sentencing
comments in their case, most thought jurors would be interested. They responded as
follows:

Very interested - 64%

Quite interested - 34%

Not at all interested - 2%
Participants were also asked about the usefulness of the booklet in forming their
judgment about the appropriateness of the sentence (B8(a)). Most found it useful:
60
Part 3: Results
Very useful - 37%; Fairly useful - 53%; Not very useful - 9%; and Not at all useful 1%. More than 90 percent of respondents also stated that knowing the reasons the
judge gave for sentence affected their view of the appropriate sentence.
Figure 21 cross-tabulates responses on the juror‟s view of the appropriateness of the
judge‟s sentence with the impact of the sentencing comments. As shown, the more
appropriate the respondent thought the sentence was, the more impact the sentencing
comments had. Nearly two thirds who said the judge‟s sentence was very appropriate
also reported that the sentencing comments impacted a lot.
Figure 21: Impact of sentencing comments
Percentage
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Very
appropriate
n=199
Fairly
appropriate
n=192
Inappropriate
n=44
Total
N=435
A lot
63.8%
45.8%
15.9%
51.0%
A little
31.2%
46.4%
63.6%
41.1%
Nothing
5.0%
7.8%
2.5%
7.8%
Participants were also asked to estimate the effect of the information on crime levels
and trends on their judgment of the appropriateness of the sentence. The relationship
between these two items was less definitive with only 74 jurors (18%) saying it
affected their judgement „a lot‟. Most thought it affected their judgement on the
sentence „a little‟ (50%) or „not at all‟ (32%).
6. Jurors’ general opinion of current sentencing practice after sentence
In Questionnaire 2, after they had been informed of the sentence and received the
booklet, participants were again asked whether, in general, sentences across a range of
offence types were too tough or too lenient (using a five-point Likert scale).
Comparing the responses to the item in Questionnaire 1 with those to the identical
item in Questionnaire 2 reveals a number of interesting results:

„Too lenient‟ responses decreased for all crime categories.

This applies to „much too lenient‟ and „a little too lenient‟ responses as well.

As a result „about right‟ responses increased across all crime categories.

Interestingly, too tough responses increased for drug offences (to 10
percent).

The most common response across all offence categories remained „too
lenient‟ except for property offences for which the most common response
was „about right‟.

„Much too lenient‟ was no longer the most common response for sex
offences.

At Stage 2, around two thirds considered sentences for sex and violent
offences were „too lenient‟.
61
Jury Sentencing Survey
The following tables illustrate these findings by comparing Questionnaire 1
responses, first with all Questionnaire 1 respondents, and then with Stage 2
respondents only (omitting those who did not go on to Stage 2 of the study).
Table 29: Jurors’ views on sentencing Stage 1 and Stage 2 compared
Type of Offence
Juror’s opinion (all respondents)
Sex
Violent
Drugs
Property
Q1
Q2
Q1
Q2
Q1
Q2
Q1
Q2
Much too lenient
41
33
23
18
21
20
15
12
A little too lenient
39
37
53
48
35
29
39
34
About right
18
29
23
33
36
41
42
50
A little too tough
1
1
1
1
6
8
4
3
Much too tough
1
0
0
0
Q1: Violent N=681, Property N=674, Drugs N=677, Sex N=674.
Q2: Violent N=434, Property N=432, Drugs N=431, Sex N=429.
2
2
0
0
Table 30: Jurors’ views as to sentencing in general, Stage 1 and Stage 2
compared (Stage 2 respondents only)
Type of Offence
Juror’s opinion – Respondents
who completed Q1 and Q2
a
Sex
Violent
Drugs
Property
Q1
Q2
Q1
Q2
Q1
Q2
Q1
Q2
Much too lenient
40
33
23
18
21
20
15
12
A little too lenient
38
37
50
48
33
29
39
34
About right
20
29
26
33
37
41
42
50
A little too tough
1
1
1
1
7
8
3
3
0
0
2
2
0
0
Much too tough
1
0
Due to rounding percentages do not equal 100%.
a
Comparing the results for Stage 1 respondents (excluding those who completed
Questionnaire 1 only) with the responses for all Stage 1 respondents showed a very
similar pattern. However, Stage 2 respondents as a group were a little less likely in
Stage 1 to say sex, violent and drug offences were too lenient, more likely to say they
were about right and more likely to say drug offences and property offences were too
tough. This raises a question about those respondents who only completed
Questionnaire 1. The following table shows the responses for those respondents only
and compares them with Stage 2 respondents.
Table 31: Jurors’ views as to sentencing in general, Questionnaire 1 responses
compared for Stage 1 only and Stage 2 respondents
Questionnaire 1 –
responses of those
who only completed
Questionnaire 1 and
those who completed
Questionnaire 1 & 2
Too lenient
62
Type of Offence
Sex
Violent
Drugs
Property
Q1
only
Q1 &
Q2
Q1
only
Q1&
Q2
Q1
only
Q1 &
Q2
Q1
only
Q1 &
Q2
84
78
81
73
61
54
54
54
About right
15
20
18
26
32
37
41
42
Too tough
1
2
1
1
7
9
5
4
Part 3: Results
These results suggest that Stage 2 respondents were less punitive at Stage 1 than those
who completed Stage 1 only. For all offence types other than property offences the
Stage 1 only participants were more likely to say sentences for sex, violence and drug
offences were too lenient and less likely to say they were about right.
The important point from the analysis of changes in jurors‟ opinions of sentencing
levels is that in Stage 2, after learning of the sentence imposed and receiving the
Crime and Sentencing booklet, respondents were less likely to say sentences were too
lenient. However, in Stage 2, 70 percent and 66 percent of respondents still
considered that sentences for sex and violent offences were too lenient. This is so
even though 50 percent of respondents in violent offence cases said the sentence was
very appropriate and 51 percent had suggested a more lenient sentence than the judge.
For property offences 46 percent said sentences were too lenient but only 30 percent
had suggested a more severe sentence than the judge. Interestingly, there was no such
clear dichotomy in the case of drug offences with 49 percent saying sentences were
too lenient and just 46 percent preferring a more severe sentence at Stage 2.
Do jurors’ opinions differ depending on the crime type of their trial?
The obvious question is whether juror‟s opinions differ depending on the crime type
of the trial that they themselves participated in. Table 32 shows some interesting
results. The fact that jurors were on a sex offence trial did not appear to have any
impact on their general view about sex offender sentencing. They were just as likely
as other jurors to say sentences for sex offences were too lenient. Similarly, for drug
offences, the responses of the drug trial jurors to sentencing patterns for drug offences
were almost identical to the pattern of responses for all jurors. However, jurors on
violent offence trials were more likely to say sentences for violent offences were
about right and less likely to say they were too lenient than other jurors. And property
offence jurors were more likely to say that sentences for property offences were about
right and less likely to say they were too lenient than other juror respondents.
Table 32: General perceptions of sentencing leniency by respondent’s trial type
Juror trial type (Q2 respondents only)
Too tough
About right
%
%
Sex offence N = 428
Too lenient
%
Total
%
Jurors on sexual offence trial n = 89
1
29
70
100
Other jurors
1
29
70
100
Total
1
29
70
429
Violent offence (not sexual) N = 434
Jurors on violence trial n = 151
1
37
62
100
Other jurors
1
31
68
100
Total
1
33
66
100
Drug offence N = 431
Jurors on drug trial n = 90
11
41
48
100
Other jurors
10
41
49
100
Total
10
41
49
100
Property offence N = 432
Jurors on property trial n = 67
3
61
36
100
Other jurors
4
49
47
100
Total
4
50
46
100
63
Jury Sentencing Survey
Why does opinion in relation to sentencing leniency in general change? Knowing the
sentence (in many cases it was more severe than the juror‟s sentencing choice) and
knowledge of sentencing patterns from the booklet may be important in changing
general views. One could hypothesise that this would have a greater impact in relation
to the type of crime that the juror tried.
This association holds for violent and property offences. Violent and property offence
jurors were more likely to say sentences in general were about right and less likely to
say they were too lenient than other jurors. The opinions of property offence jurors as
to the sentence choice and their general attitudes to sentencing leniency for property
offences also appear to be quite consistent with 31 percent suggesting a more severe
sentence at Stage 1 (see Figure 9) and 36 percent stating sentences for property
offences were too lenient. This is not the case, however, for violent offence jurors
with 49 percent suggesting a more severe sentence but 62 percent of jurors saying
sentences were too lenient. In the case of sex and drugs, while opinions became more
lenient overall, the fact that jurors had experienced a trial and sentencing for a sex
offence or a drug offence did not seem to affect their general attitudes to sex or drug
offences more than other jurors. For sex offences, even though only 45 percent
suggested a more severe sentence than the judge (see Figure 9), 70 percent of sex
offence jurors said sentences or sex offences were too lenient (the same proportion of
other jurors who had this view). So, in summary, the type of crime tried by the juror
themselves only seemed to have an impact on attitude change for violent and property
offences.
Changing Levels of Punitiveness?
The next question is whether those who selected a more lenient sentence at Stage 1
were less likely to have said sentences were too lenient at Stage 2 than those who
selected a more severe sentence. To measure this, a second general punitiveness index
was created from juror responses as to whether sentences were too tough, about right
or too lenient across the four offence categories (the Stage 2 general punitiveness
index). This variable showed that those who had selected a more severe sentence than
the judge at Stage 1 had statistically higher mean scores on the Stage 2 general
punitiveness index than those who had chosen a less severe sentence than the judge
(p=.01). So while those who selected a more lenient sentence than the judge in Stage
1 are less likely to have said sentences are too lenient at Stage 2 than those who
selected a more severe sentence, as the discussion of Table 32 suggests, there are
clearly some sex and violence offence jurors, in particular, who selected a more
lenient sentence than the judge but who have persisted with the view that sentences, in
general, are too lenient. We have also seen that, except for property offences, the
offence type of trial has only a marginal impact on the general question about
sentencing severity.
7. Changes in jurors’ knowledge of crime and sentencing trends
The questions about crime trends and sentencing patterns were repeated in
Questionnaire 2 after jurors had received the booklet, which contained information on
each of the questions asked within the surveys. As shown in the Table below, while
the number of accurate responses increased to 50 percent (combining decreased a
little and a lot), a significant proportion of jurors (38%) still thought that recorded
crime had increased, with 11 percent saying that it had increased a lot. So even though
64
Part 3: Results
62 percent of respondents stated that they read the booklet in full, only 50 percent
used that information to develop a more accurate response to the trends in crime
questions.
Table 33: Changes in jurors’ perceptions of overall recorded crime trends
Increased a
lot
Increased a
little
Stayed
same
Decreased a
little
Decreased a
lot
Don’t
know
26
37
15
6
1
15
11
27
10
31
19
2
Q1
N=442
Q2
N=438
However, our data do indicate that greater access to information can make a
difference to juror attitudes. As shown in Table 34, those who read the booklet in full
were statistically significantly more likely to answer the question in relation to crime
trends accurately compared with those who read some sections in full or those who
said they flicked through it (p=.000).
Table 34: Reading the booklet and assessment of overall crime trends
Read the
booklet in
full
%
Assessment of
general crime
trends
In relation to the booklet did the juror…?
Read
Flick
some
Not read
through
sections in
the booklet
the booklet
full
%
%
%
44.0
23.6
50.0
Total
%
accurate
56.8
49.5
inaccurate
43.2
56.0
76.4
50.0
50.5
n=
271
100
55
8
434
Chi Square Test statistically significant p=.000
Jurors‟ levels of knowledge improved across all offence categories between
Questionnaire 1 and Questionnaire 2. However, as indicated in Table 35, below, a
significant proportion still believed that crime for burglary, robbery, rape and motor
vehicle theft is rising. For example, 30 percent of respondents responded that motor
vehicle theft was becoming more common even though the booklet explained quite
clearly that the general trend was downwards.
Table 35: Changes in jurors’ perceptions of crime trends for selected offences
Juror’s opinion –
Respondents who
completed Q1 and Q2
Type of Offence
Burglary
Robbery
Rape
MV Theft
Murder
Q1
N439
Q2
N437
Q1
N439
Q2
N434
Q1
N436
Q2
N434
Q1
N441
Q2
N435
Q1
N438
Q1
N434
More Common %
58.8
20.8
56.0
23.7
15.8
17.3
57.6
29.9
15.8
6.9
Stayed about the same %
27.3
23.8
32.1
39.9
47.9
54.1
26.8
20.5
47.7
66.6
Less common %
5.9
52.6
4.3
33.9
17
18.2
6.6
47.4
21.2
18.7
Don’t know %
8.0
2.7
7.5
2.5
19.3
10.4
9.1
2.3
15.3
7.8
As shown in Table 36, this pattern was also evident in jurors‟ beliefs about the level
of crime that involved violence. Knowledge improved, however, a significant
proportion of the sample were still inaccurate in their estimation of violent crime
despite the booklet details.
65
Jury Sentencing Survey
Table 36: Changes in perception of proportion of crime that involves violence
19.6
Between 1/4
and 1/2
32.5
Between 1/2
and 3/4
28.4
62.7
17.7
8.8
Questionnaire
1/4 or less
Q1 N=443
Q2 N=434
More than 3/4
Don’t know
12.0
7.4
4.1
6.7
The obvious question from these data is whether the continuation of distorted views
of crime is related to the level of attention given to the information booklet by the
juror. Table 37 correlates these responses with the responses in relation to how well
the booklet was read. As can be seen, those who read the booklet in full were much
more likely to answer the question in relation to the proportion of crime that involves
violence more accurately compared with those who only read some sections in full
(p=.000). Persistent misconceptions about crime may therefore be explained in least
in part by lack of reliable information.
Table 37: Reading the booklet and correct assessment of the proportion of crime
that involves violence
In relation to the booklet did the juror…?
Read the booklet
in full
%
Assessment of accurate
72.7
proportion of
crime that
inaccurate
27.3
involved
n=
(271)
violence
Chi Square Test statistically significant p=.000
Read some
sections in full
%
Flick through
the booklet
%
Not read the
booklet
%
Total
%
53.5
32.1
37.5
62.6
46.5
67.9
62.5
37.4
(99)
(53)
(8)
(431)
The pattern displayed in previous data in this section continued on the items relating
to perceptions of crime rates and sentencing. As shown in the table below, while the
proportion of respondents with an accurate answer about imprisonment rates for rape
increased to 65 percent and a greater improvement was seen in relation to the burglary
imprisonment rate perceptions, almost half of the participants in Stage 2 still
underestimated the imprisonment rate for burglary.
Table 38: Changes in perception of imprisonment rates
Crime
Burglary
Rape
a
Questionnaire
Perception of imprisonment rate
0-25%
26- 50%
51-75%
75% +
Q 1 N=436
37.6
41.1
18.1
a
Q 2 N=417
23.7
25.4
43.9
a
7.0
Q 1 N=432
9.7
27.8
30.6
31.2
a
Q 2 N=418
8.6
12.9
13.4
65.1
a
3.2
Indicates accurate response.
Further analysis of these data showed that there was an inter-correlation between the
answers to the questions on imprisonment rates. Those who answered the burglary
imprisonment rate accurately were more likely to answer the rape imprisonment
question accurately and vice versa. For example, nearly 90 percent who answered the
burglary imprisonment rate accurately answered the rape rate accurately compared
with 65 percent overall who were accurate.
66
Part 3: Results
Table 39: Use of the booklet and knowledge of burglary imprisonment rate
In relation to the booklet did the juror…?
Assessment of
burglary
imprisonment rate
Assessment of rape
imprisonment rate
Read the
booklet in
full
%
Read some
sections in
full
%
Flick
through the
booklet
%
Not read
the booklet
%
Total
%
50.4
41.7
22.4
0.0
44.3
inaccurate
49.6
58.3
77.6
100.0
55.7
n=
(262)
(96)
(49)
(6)
(413)
accurate
73.0
55.2
44.9
33.3
65.0
inaccurate
27.0
44.8
55.1
66.7
35.0
accurate
n=
(263)
(96)
(49)
(6)
(414)
Chi Square Test indicates that differences between different attention paid to the booklet are statistically
significant p=.000.
Figure 18 (above) showed that almost two thirds read the booklet in full, almost a
quarter read some sections in full and nine percent flicked through it. Table 38 shows
that in Stage 2, 44 percent of respondents accurately answered the question about
burglary imprisonment rates. Table 34 shows a correlation between those who read
the booklet and knowledge of crime trends. As expected, Table 39 shows a similar
correlation between use of the booklet and knowledge of imprisonment rates.
There were no significant differences in gender, age, income or marital status between
those who answered the imprisonment rate questions accurately and those who did
not. However, as Table 40 shows there were significant differences between
education levels of those who answered the questions correctly – 77 percent of those
with a bachelor degree or above answered the rape question correctly compared with
47 percent of those educated to Year 10 or lower.
Table 40: Knowledge of imprisonment rates and education level of respondent
Highest level of education completed
Knowledge of
imprisonment
rate for
burglary
Knowledge of
imprisonment
rate for rape
Year 10 or
below
%
Year 11
%
Year 12
%
Trade or
Apprentice
ship
%
Certificate
or Diploma
%
Bachelor's
degree or
above
%
Total
%
accurate
33.3
23.1
44.4
33.3
48.1
52.8
43.8
(182)
inaccurate
66.7
76.9
55.6
66.7
51.9
47.2
56.3
(234)
n=
(78)
(13)
(45)
(51)
(104)
(125)
(416)
accurate
47.4
53.8
57.8
69.2
66.3
76.8
65.0
(271)
inaccurate
52.6
46.2
42.2
30.8
33.7
23.2
35.0
(146)
n=
(78)
(13)
(45)
(52)
(104)
(125)
(417)
67
Jury Sentencing Survey
8. Changes in jurors’ perceptions of the risk of victimisation and safety
Questionnaire 2 also repeated the question on juror‟s estimation of their own risk of
victimisation for burglary, motor vehicle stealing, assault and robbery. Table 41
shows that after receiving the booklet, estimations of the risk of victimisation over the
next 12 months were more accurate but still only about half of the participants
managed to correctly estimate the risks for burglary (having your home, garage or
shed broken into) and motor vehicle theft.
Table 41: Estimation of the risk of victimisation for various offences, percent.
Estimation of risk
Crime
Burglary
MV theft
Assault
Robbery
Questionnaire
Less than
6%
6-10%
11-30%
31-50%
50% plus
Q1 N = 440
33.2
32.0
17.0
12.5
5.2
Q2 N = 440
49.5
25.5
11.1
10.5
3.4
Q1 N = 442
39.4
28.1
18.1
10.4
4.1
Q2 N = 439
55.8
22.3
13.0
6.2
2.7
Q1 N = 438
57.3
26.5
10.7
3.9
1.6
Q2 N = 440
67.0
17.3
10.7
3.6
1.4
Q1 N = 439
43.3
33.0
15.5
7.3
.9
Q2 N = 440
62.0
19.3
12.3
4.3
2.0
The perceptions of safety question from Questionnaire 1 was also repeated in
Questionnaire 2. As shown in Table 42 below, there was some, but limited change in
juror‟s perception of their own safety, from less safe to more safe, after receiving the
booklet.
Table 42: Perceptions of safety
Questionnaire
Very safe
%
Fairly safe
%
A bit unsafe
%
Very unsafe
%
Walking alone
after dark
Q1
20.9
45.3
28.6
5.2
Q2
26.4
48.0
20.5
5.2
At home
alone at night
Q1
48.6
44.6
6.1
.7
Q2
52.7
41.2
4.5
1.6
How safe
when:
N = 444
9. Jurors’ opinions of judges after sentence (Stage 2)
In Questionnaire 2, participating jurors were again asked whether judges were in
touch about public opinion on sentencing. Figure 22 shows that in Stage 2 the
proportion of those who responded that judges were very in touch doubled from 13
percent to 26 percent, the proportion of those who responded somewhat in touch was
the same and the „somewhat out of touch‟ responses decreased. The data was run
using Questionnaire 2 respondents only. The pattern of responses was very similar
between all Questionnaire 1 respondents and the Stage 2 participants‟ answers to
Questionnaire 1. In other words, the differences between Questionnaire 1 and
Questionnaire 2 responses were not due to the fact that the two groups were
differently constituted. It can be concluded, therefore, that the change in jurors‟
perceptions of judge‟s level of being in touch with public opinion is associated with
knowledge of the sentence and the information received.
68
Part 3: Results
Figure 22: How in touch are judges, changes in response (all respondents)
60.0%
Percentage
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Very in touch
Somewhat in
touch
Somewhat out
of touch
Very out of
touch
Q1 N=688
12.6%
57.0%
28.1%
2.3%
Q2 N=440
25.7%
56.8%
15.0%
2.5%
Not surprisingly, as Table 43 shows, those who said judges were very in touch were
more likely to say the sentence was very appropriate than fairly appropriate or fairly
inappropriate. Those who said that judges were somewhat in touch or somewhat out
of touch were more likely to say the sentence was fairly appropriate than very
appropriate or fairly inappropriate.
Table 43: The Relationship between perceptions of judicial remoteness and the
appropriateness of the sentence.
How appropriate was the sentence?
How in touch are the
Judges?
Very
appropriate
%
Fairly
appropriate
%
Fairly
inappropriate
%
Very
inappropriate
%
Total
%
Very in touch
59.8
38.4
.9
.9
100 (112)
Somewhat in touch
42.7
49.6
7.3
.4
100 (248)
Somewhat out
30.3
39.4
24.2
6.1
100 (66)
Very out of touch
36.4
18.2
36.4
9.1
100 (11)
Total
45.1
44.4
8.9
1.6
100 (437)
The next question is how do views of judicial remoteness relate to general views as to
sentence and to sentence choice? Table 29 showed that in Stage 2, two thirds or more
of respondents thought sentences for sex and violent offences were too lenient and for
drug and property offences a little less than half were of this view. Table 44 shows
that for sex, violent, drug and property offences the more in touch the judge was said
to be, the less likely sentences were said to be too lenient and the more likely they
were said to be about right. For all but sex offences, respondents with the view that
judges were very in touch were more likely to say the sentence was about right than
too lenient. However, for sex offences, even those who said judges were very in touch
were more likely to say that sex offence sentences were too lenient than about right.
This relationship between whether or not judges are perceived to be in touch seems to
be a measure of punitiveness. Similarly, there appears to be a relationship between
whether or not judges are perceived to be in touch and punitiveness as measured by
sentence choice in Stage 1 (Stage 1 comparative sentence variable).
69
Jury Sentencing Survey
Table 44: Relationship between perceptions of judicial remoteness and general
attitudes to sentences (Stage 2)
Q2: Are current sentences too tough/lenient? Sex offence
How in touch are
Judges:
Very in touch
n = 111
Somewhat in touch
n = 241
Somewhat out of touch
n = 66
Very out of touch
n = 10
Total
N = 428
Too lenient
%
About right
%
Too tough
%
58.6
39.6
0.8
70.5
28.2
1.2
81.8
18.2
0.0
100.0
0.0
0.0
69.9
29.0
1.2
Q2: Are current sentences too tough/lenient? Violent offence (not sexual)
How in touch are
Too lenient
About right
Too tough
Judges:
%
%
%
Very in touch
n = 112
Somewhat in touch
n = 243
Somewhat out of touch
n = 66
Very out of touch
n = 11
Total
N = 432
48.2
50.9
0.9
67.1
31.7
1.2
86.4
13.6
0.0
100.0
0.0
0.0
66.0
33.1
0.9
Q2: Are current sentences too tough/lenient? Drug offence
Too lenient
About right
%
%
How in touch are
Judges:
Very in touch
n = 113
Somewhat in touch
n = 240
Somewhat out of touch
n = 66
Very out of touch
n = 11
Total
N = 430
Too tough
%
32.7
54.9
12.4
48.8
41.7
9.6
71.2
19.7
9.1
90.9
9.1
0.0
49.1
40.9
10.0
Q2: Are current sentences too tough/lenient? Property offence
Too lenient
About right
Too tough
%
%
%
How in touch are
judges?
Very in touch
n = 112
Somewhat in touch
n = 242
Somewhat out of touch
n = 66
Very out of touch
n = 10
Total
N = 430
31.3
63.4
5.4
44.6
51.7
3.7
69.7
27.3
3.0
80.0
20.0
0.0
45.8
50.2
4.0
The next table suggests that the more in touch judges were perceived to be the more
likely it was that respondents had chosen a less severe sentence than the judge. For
example, 57 percent of those who said judges were very in touch also selected a more
lenient sentence than the judge in Stage 1 compared with 35 percent who said judges
were somewhat out of touch. However, the results were not significant.
70
Part 3: Results
Table 45: The Relationship between perceptions of judicial remoteness and
sentence choice at Stage 1
Stage 1 comparative sentence severity
How in touch are Judges?
Very in touch
Less severe
%
56.6
Same
%
4.7
More severe
%
38.7
Total
%
100 (106)
Somewhat in touch
52.2
3.9
43.9
100 (228)
Somewhat out of touch
34.9
6.3
58.7
100 (63)
Very out of touch
55.6
.0
44.4
100 (9)
Total
50.7
4.4
44.8
100 (406)
Table 46 shows the relationship between whether judges are perceived to be in or out
of touch and sentence preference at Stage 2 (the Stage 2 comparative sentence
variable). It shows the more in touch judges were perceived to be, the less punitive
was the respondent (p=.015). For example, of those who said judges were somewhat
out of touch, 54.5 percent preferred a more severe sentence, whereas of those who
said judges were very in touch, only 24.8 percent preferred a more severe sentence.
Table 46: Relationship between perceptions of judicial remoteness and sentence
preference at Stage 2
Stage 2 comparative sentence severity
How in touch are judges?
Very in touch
Less severe
%
12.4
Same
%
62.8
More severe
%
24.8
Total
%
100 (113)
Somewhat in touch
8.9
53.6
37.5
100 (248)
Somewhat out of touch
4.5
40.9
54.5
100 (66)
Very out of touch
9.1
36.4
54.5
100 (11)
Total
9.1
53.7
37.2
100 (438)
Chi Square Test statistically significant p=.015
10. Attitudes to public opinion, punishment and law breaking
Questionnaire 2 asked jurors (C10) how much they agreed with each of four
statements. This question is duplicated from an item used in the AuSSA 2003 and
AuSSA 2007, although the statement, „People who break the law should be given
stiffer sentences‟ was excluded because that question, in a different form, had already
been asked in the question about whether sentences were too tough or too lenient
(question C1). Indermaur and Roberts (2005: 152-153) found that majority of AuSSA
2003 respondents (63%) agreed that judges should reflect public opinion about crimes
when sentencing criminals. In 2007, this dropped slightly to 58.4 percent (Roberts and
Indermaur 2009: 20).
Our data, outlined in Table 56, show a similar response although our juror
respondents were less likely to strongly agree than AuSSA 2003 respondents (20%
compared with 31%). Interestingly, in AuSSA 2007, those who had contact with a
criminal court in the past 12 months were significantly less likely to agree that judges
should reflect the views of the public when sentencing (53.6% versus 70.8%)
(Roberts and Indermaur 2009: 20).
In AuSSA 2003, almost half agreed the death penalty should be the punishment for
murder and in AuSSA 2007 the response was about 4 out of 10 in contrast with just
71
Jury Sentencing Survey
25 percent in this study. Roberts and Indermaur (2009: 18) report that in AuSSA 2003
and 2007, seven out of 10 believed that those who break the law should be given
stiffer sentences – a similar response to our Questionnaire 1 jurors‟ response to the
too tough or too lenient question in relation to sex and violent offences (see Table 19).
In 2003, those who agreed that judges should reflect public opinion were about twice
as likely to agree with stiffer sentences and that the death penalty should be the
punishment for murder (Indermaur and Roberts 2005: 153).
Table 47: Attitudes to public opinion, punishment and law breaking (percent)
Judges should reflect public opinion
when sentencing (N = 437)
Death should be the punishment for
murder (N = 440)
The law should always be obeyed
even if it is wrong (N = 441)
Breaking the law to protect a family
member or friend is sometimes
justified (N = 439)
Strongly
agree
%
Agree
%
Neither agree
nor disagree
%
Disagree
%
Strongly
disagree
%
20
45
18
14
3
10
15
24
23
28
6
33
31
25
5
9
38
29
20
5
The next Table shows that those who agree that judges should reflect public opinion
in sentencing were three times more likely to agree with the death penalty than those
who disagreed.
Table 48: Agreement with death penalty for murder by agreement that judges
should reflect public opinion in sentencing, percent
Should death penalty be punishment for murder
Should
judges
reflect
public
opinion
Yes
%
Neither agree or
disagree
%
No
%
Judges should reflect
public opinion (N = 284)
30
27
43
Neither agree or disagree
(N = 79)
20
28
52
Judges should not reflect
public opinion (N = 72)
10
8
82
Were our respondents who agreed that judges should reflect public opinion more
punitive on other measures? We have derived measures of punitiveness from
comparing the juror‟s sentence preference with the judge‟s sentence and from jurors‟
general views as to the severity of sentencing. Table 49 looks at the Stage 1
comparative sentence variable (the comparison between the judge and juror‟s
sentence).
Table 49: Agreement that judges should reflect public opinion in sentencing by
choice of more or less severe sentence at Stage 1
Judges should reflect
public opinion
Agree n= 291
72
Stage 1 comparative sentence severity
More severe
Same
Less severe
%
%
%
47
4
49
Neutral n= 77
48
6
45
Disagree n=79
66
2
32
Total N = 447
51
4
45
Part 3: Results
In Stage 1, 51 percent of Stage 2 jurors were more lenient than the judge and 45
percent were more severe (see Figure 9). While those who agreed that judges should
reflect public opinion were quite evenly split between those who selected a more
severe and a less severe sentence, those who disagreed that judges should reflect
public opinion were half as likely to have chosen a more severe sentence than the
judge. While these results are not statistically significant (p=.099) there is a trend
suggesting that, on this measure, those who favour the judge taking public opinion
into account are more punitive than the judges.
The next table uses the Stage 2 comparative sentence severity variable. At Stage 2, of
those who answered the question about judges‟ responsiveness to public opinion, only
nine percent wanted a more lenient sentence, 53 percent agreed with the judge‟s
sentence and 38 percent wanted a more severe sentence. Table 50 shows that
respondents who agreed judges should reflect public opinion in sentencing were more
likely than those who disagreed with the statement to have preferred a more severe
sentence than the judge.
Table 50: Agreement that judges should reflect public opinion in sentencing by
sentence preference at Stage 2.
Judges should reflect public
opinion in sentencing
Agree
n = 283
Neither agree nor disagree
n = 80
Disagree
n = 72
Total
N = 435
Stage 1 comparative sentence severity
Less severe
Same
More severe
%
%
%
9.2
50.5
40.3
6.3
57.5
36.3
12.5
58.3
29.2
9.2
53.1
37.7
Using the Stage 2 general punitiveness index, the mean punitiveness scores of
respondents were compared by how strongly they agreed or disagreed with the
statement that judges should reflect public opinion in sentencing. The results indicate
that there was a positive linear association between the respondents‟ punitiveness
score and their level of agreement with the statement. The mean score of those who
strongly agreed with the statement (n = 85) was 16.2 (in a range of 7-20) compared
with a mean score of 12.9 (n = 12) for those who strongly disagreed with the
statement. The difference in mean punitiveness score by the public opinion item was
statistically significant (p = .000). In summary, those agreeing that judges should
reflect public opinion were more punitive on each of our three measures of
punitiveness, namely, Stage 1 comparative sentence severity, Stage 2 comparative
sentence severity and the Stage 2 general punitiveness index.
The statement, „Judges should reflect public opinion about crimes when sentencing
criminals‟ is also related to the idea of whether judges are in touch.
73
Jury Sentencing Survey
Table 51: Agreement that judges should reflect public opinion in sentencing by
how in touch judges are with public opinion, percent
Judges should reflect
public opinion
Are judges in or out of touch with public opinion on sentencing
Very in touch
Somewhat in touch
Out of touch
%
%
%
Agree
n = 283
Neutral
n = 79
Disagree
n = 71
Total
N 433
21
58
21
32
56
13
41
49
10
26
56
18
Chi Square Test statistically significant p=.000
Whether respondents agreed, were neutral or disagreed that judges should reflect
public opinion, they most commonly responded that judges were somewhat in touch.
However, those who agreed that judges should reflect public opinion were more likely
than those who were neutral or disagree to say judges were out of touch and those
who disagreed were more likely than those who agree or were neutral to say judges
were very in touch. This finding was statistically significant (p=.000).
11. Did jurors discuss the sentence or study with family or friends?
The final question asked jurors participating in Stage 2, if they had discussed the
sentence, the information on crime trends or sentencing patterns with family or
friends. This was to assess if jurors could operate as conduits of public opinion about
sentencing in a way that might help improve knowledge of crime and sentencing
among the general public. The results show that:

68% discussed the sentence

33% discussed the information on crime trends

28% discussed the information on sentencing trends
The finding that 68 percent discussed the sentence is an important one in the light of
the fact that 90 percent of Stage 2 respondents thought the sentence was appropriate.
The following table shows that those who would have preferred a more lenient
sentence and those who agreed with the judge‟s sentence were a little more likely to
have discussed the sentence with family and friends than those who would have
preferred a more severe sentence, although the result was not statistically significant.
However, it is interesting that it is not just those who feel a sentence was too lenient
who are likely to discuss the sentence with others.
Table 52: Whether discussed the sentence by sentence preference at Stage 2
Have your discussed with family and friends: The
sentence
No
Yes
Total
%
%
%
Stage 2 comparative sentence
severity
74
Less
severe
22.5
77.5
100 (40)
Same
31.5
68.5
100 (235)
More
severe
35.5
64.5
100 (166)
Total
32.2
67.8
100 (441)
PART 4
DISCUSSION
A. MAJOR FINDINGS
In this Part of the report the major findings are discussed, the research questions
posed at the outset of the project are addressed and policy implications are explored.
1. Jurors are willing to be used as a source of public opinion
The response rate of 35 percent of jurors for Questionnaire 1 suggests that at least a
third of jurors are willing to participate in a sentencing survey designed to elicit
public opinion about sentencing. This result is broadly similar to the Western
Australian Jury Intimidation Project, which asked jurors to fill out a 24 page survey
form and which attracted a response rate of 33 percent (Fordham 2009). Once jurors
had agreed to participate in Stage 1, the take up rate for Stage 2 was quite strong with
64 percent of Stage 1 respondents completing Stage 2. Moreover, almost half of the
Stage 2 respondents were willing to be interviewed (see Table 24). The key to a good
response rate is capturing jurors at Stage 1.
The degree of interest a particular case aroused, rather than offence seriousness,
appeared to be the critical factor in a achieving a good response rate. For example, a
high response rate (ten jurors) was achieved in an early assault trial where the
outcome was conditional release without a conviction. And in another assault case,
which attracted a wholly suspended sentence of three months, 11 jurors participated.
In the same way that offence seriousness did not appear important as a determinant of
response rate, neither did the type of crime (sex, violence, drugs, property or other),
the length of trial, the time taken to deliberate nor the adjournment of sentencing
submissions. However, both the presiding judge and the place of trial had an effect.
Discounting the response rate of 92 percent for one judge who presided over only one
trial, response rates varied between judges from 44 percent down to 27 percent. Quite
why the judge was a relevant factor is not clear. It may be that the judge‟s perceived
interest in and endorsement of the project was important. This is supported by the fact
that eleven jurors responded in the only trial presided over by Judge 1, who was the
judge who initially approved the project, but who subsequently retired from the Court.
Place of trial was a relevant factor attributed, at least in part, to different standards of
jury facilities and perhaps to differences in support and commitment of court staff.
The lower response rate in Burnie, 14 percent compared with 46 percent in Hobart,
suggests that it may be difficult to obtain a high response rate in small cities or
centres. Jurors in such places may feel less anonymous, and may not want to be
associated with the sentencing process by responding to a request to stay to listen to
the sentencing proceedings. The higher response rates that have been achieved in
some other jury studies and the relatively strong response rate in Hobart in this study
suggest that there is potential to improve the response rate. A shorter time frame,
which would be possible in jurisdictions with more trials per month, would also assist
in improving participation.
75
Jury Sentencing Survey
2. Jurors are reasonably representative of the general population
Juror respondents were found to be reasonably representative of the Tasmanian
population in gender and age with a slight over-representation of respondents in the
45-64 age group and an under-representation of those of 65 and over. As anticipated,
Australian born respondents were over-represented because of the requirements of
enrolment on the jury panel. Juror respondents were less likely to be unemployed than
the general population and more likely to be better educated and have a higher income
than the general population. However, in terms of current occupation, they were
closely representative of the jury-aged population.
3. Informed public opinion is not as punitive as the populist view of public
opinion suggests
The finding that jurors were slightly more likely to be more lenient than the judge
rather than more severe in their sentence choice at Stage 1 (52 percent were more
lenient and 44 percent were more severe) contrasts strikingly with the findings in
representative surveys which indicate that about 70 percent of the public think that
sentences are too lenient (Gelb 2006: 11) or that those breaking the law should
receive stiffer sentences (Indermaur and Roberts 2009: 18). Instead, it accords more
with the findings of studies which have compared judicial sentences with those
selected by members of the public by using vignettes (Diamond and Stalans 1989;
Lovegrove 2007). Diamond and Stalans (1989), who surveyed persons presenting for
jury duty and university students in Canada, found the judges‟ sentences were as
severe or more severe than those of the members of the public. In Lovegrove‟s study,
a majority of respondents selected sentences less than the judges for theft (71%)
aggravated robbery (86%) and rape (63%). For the fourth case, intentionally causing
serious injury, a majority were more severe than the court (65%) (Lovegrove 2007:
776). Our findings, like those of Diamond and Stalans and Lovegrove, contrast with
those of De Keijser, Koppen and Elfers (2007) who found that the choice of sentence
of members of the public who had been given the same case study as judges, were not
as harsh as those given shorter accounts but were still significantly harsher than the
judge.
There are important differences between Lovegrove‟s Victorian study and our study
which can explain why more of our participants in this study were more severe than
the judge compared with Lovegrove‟s sample. The first difference lies in the timing
and process of informing the survey participants. Lovegrove‟s participants first
received a 70 minute lecture about sentencing to give them an idea of the law‟s sense
of justice and then the judge gave them data on current sentencing practice before
they selected the sentence. In contrast, our juror respondents selected the sentence
after knowing only the facts about the offence and the offender and without being
briefed about sentencing or current sentencing levels. The second difference relates to
the nature and variety of the cases that were used: Lovegrove‟s four test cases were all
selected so that the offenders had „potentially strong claims to mitigation‟ (Lovegrove
2007: 773), whereas none of the cases in our study were selectively chosen. Every
case was included simply on the basis that it had been held in the Supreme Court of
Tasmania during the survey period. Another important difference is that rather than
using a small number of different cases (Lovegrove had four cases, a theft, an
aggravated robbery, a rape and a case of intentionally causing serious injury) our
76
Part 4: Discussion
study had 138 cases which were an assortment sex, violence, drugs and property
trials. This method is therefore better able to pick up broad differences between
offence types than the standard vignette methodology. Our results show a striking
difference between types of offence. For property offences, juror respondents were
more than twice as likely to be more lenient than the judge rather than more severe,
with 68 percent suggesting a more lenient sentence and 30 percent a more severe
sentence. For sex, violent and drug offences the split between less and more severe
was much more even. Respondents in culpable driving cases were also much more
likely be more lenient than the judge. However, with just three culpable driving trials
and 11 respondents, little can be drawn from this finding.
This difference between offence types was born out in Stage 2. When respondents
were asked how appropriate the sentence was after receiving the sentencing
comments and the Crime and Sentencing booklet, they were most satisfied with
sentences in property offence cases and least satisfied with sex and drug offence
sentences. Perhaps surprisingly, half of respondents thought sentences for violent
offences were very appropriate and 42 percent said they were fairly appropriate.
Comparing the judge‟s sentence with the respondent‟s preferred sentence at Stage 2
(which we called the Stage 2 comparative sentencing variable) also showed a
difference between offence types. While a majority were happy with the same
sentence as the judge in property and violent offence cases, a significant proportion
(but less than half) wanted a more severe sentence in sex and drug offence cases.
Stage 2 results provide strong support for the finding that informed public opinion is
not as punitive as general questions in representative surveys suggest with 90 percent
of respondents stating that the sentence was appropriate, evenly split between very
and fairly appropriate and only 38 percent preferring a more severe sentence than the
judge.
An interesting finding in Stage 2 is that those who were more lenient than the judge at
Stage 1 were significantly more likely to agree with the judge‟s sentence than those
who had selected a more severe sentence. This finding was replicated using the Stage
1 comparative sentence variable and the responses to the question about the
appropriateness of the sentence. So, those who selected a more lenient sentence than
the judge in Stage 1 were more likely to respond in Stage 2 that the sentence was very
appropriate and less likely to say it was inappropriate than those who had selected a
tougher sentence. Similarly, those whose sentence was more lenient than the judge‟s
sentence at Stage 1 were less likely to still want a more lenient sentence (18%)
compared with those who had chosen a more severe sentence at Stage 1 but who still
wanted a more severe sentence at Stage 2 (57%). In other words, the more punitive
respondents were less tolerant of the judge‟s sentence and were less malleable in their
views than those who were less punitive as measured by sentence choice at Stage 1.
This finding has relevance in terms of assessing the impact of information on attitude
change and helps to explain why it is that general attitudes seem to favour tougher
sentences. Those who may tend to leniency are nevertheless content with sentences
that are tougher, but those who tend to want a more severe sentence in an individual
case are less tolerant of lighter sentences.
77
Jury Sentencing Survey
4. Members of the public are more punitive when punitiveness is measured
by answers to abstract questions about sentencing than when asked about
a sentence in a particular case
As reported above, when responding to the stimulus provided by an individual case,
52 percent of respondents in Stage 1 chose a sentence that was more lenient than the
judge‟s sentence and 44 percent a more severe sentence. For property offences only
31 percent proposed a more severe and yet in response to the general question about
sentencing 54 percent said sentences were too lenient. For sex offences and violent
offences, the contrast was rather more marked: almost 50 percent had chosen
sentences which were less severe than the judge but in their response to the general
question, 80 percent said that sentences for sex offences were too lenient and 76
percent said that sentences for violent offences were too lenient. This dichotomy or
gap diminished somewhat in Stage 2 but was even more striking at this second Stage
because respondents knew the judge‟s sentence and so were aware that in many cases
it was more lenient than their selected sentence. And yet for property offences (where
only 31 percent had chosen a more severe sentence at Stage 1), at Stage 2, 46 percent
still said sentences were too lenient. For violent offences (where around half selected
a more severe sentence than the judge at Stage 1) at Stage 2 two thirds still said
sentences for violent offences were too lenient, although 50 percent of jurors with a
violent offence case said it was very appropriate and only 35 percent would have
preferred a more severe sentence. Similarly, for sex offences (where 45 percent
selected a more severe sentence at Stage 1) there were still 70 percent of respondents
who said that sentences for sex offences were too lenient although less than half
would have preferred a more severe sentence. For drug offences the gap was less and
all but disappeared in Stage 2. At Stage 1, 50 percent had chosen a more severe
sentence and 54 percent said that sentences for drug offences were too lenient. At
Stage 2, just 49 percent said sentences were too lenient for drug offences and 46
percent would have preferred a more severe sentence. (See Figure 9, Figure 13 and
Table 30).
The data were analysed to see if the dichotomy was still apparent when the general
views of respondents were separated into the types of offence tried. This was because
while respondents gave their general views for each offence category, they
necessarily only selected a sentence for one offence type in their own sentencing
exercise. For example, what did jurors who sat on a sex case think of sentencing for
sex offences at Stage 2? The data revealed that there was no difference between sex
offence jurors and other jurors, the perception gap or dichotomy remained the same
with 70 percent of sex offence jurors saying sentences for sex offences were too
lenient. For violent offence jurors the gap diminished but remained. Violent offence
jurors were less likely to say sentences for violent offences were too lenient, but still
62 percent thought sentences for violent offences were too lenient (see Table 32) even
though only 35 percent wanted a more severe sentence at Stage 2 (see Figure 13). For
property offence jurors the gap almost disappeared. Jurors were less likely to say
sentences for property offences were too lenient than other jurors with only 36 percent
saying sentences of property offences were too lenient (and 31 percent had selected a
more severe sentence at Stage 1). In summary, there is a clear contrast between
responses to general abstract sentencing views and responses to the stimulus provided
by an individual case for all offence categories except drug offences. Once
78
Part 4: Discussion
respondents became more informed this gap all but disappeared for property offences,
remained for sex offences and narrowed but remained for violent offences.
This dichotomy has been observed with social surveys in other contexts. There is a
„perception gap‟ or disparity between personal experience and societal views. A
British survey showed that 81 percent of respondents said that they were happy with
their last visit to hospital, but only 47 percent were able to say that the National
Health Service was providing a good service nationally (Taylor 2008). And the
perception gap is not restricted to assessments of government services. In a BBC poll,
93 percent of respondents described themselves as optimistic about their own family
life, yet 70 percent believe families are becoming less successful overall (Taylor
2008). In relation to confidence in criminal justice, confidence levels have been
shown to differ depending on whether the focus is local or national. In a Home Office
study, people have been found to be generally more confident with the way that crime
is being dealt with locally, than nationally (Page, Wake and Ames 2004). The same
applies to crime rates, where a decreasing proportion of people believe that crime is
increasing locally but an increasing proportion believe that it is increasing nationally
(Thorpe and Hall 2009: 66).
Our study is not the first to observe this disjuncture between responses to the abstract
question „are sentences too tough, about right or too lenient‟ and views about a
specific case. In the Diamond and Stalans (1989) study, which used vignettes to
compare the sentence choice of judges with lay respondents, the lay respondents were
not generally more severe than the judges, but two thirds of them said that judges
were too lenient in response to the abstract question. Similarly, in the Scottish Justice
1 Committee study, which used representative surveys, focus groups and a „civic
participation event‟, the responses to the general abstract question (70 percent saying
sentences were too lenient) contrasted with sentence choices in sentencing scenarios
which were not too far away from the sentence likely be imposed by a judge.
Explaining the perception gap
Varying explanations for the differences between responses to the abstract question
and to the stimulus presented by an individual case have been suggested. Diamond
and Stalans (1989: 87) focus their explanation on the way that social judgments are
made. An abstract question about judicial leniency requires a respondent to formulate
a picture of what the courts do and then to evaluate it. This picture will be based on a
knowledge of crime and sentencing that is distorted first by the media, which
selectively reports crime, and then by a biased process of recall, which ensures people
remember the violent and the extreme rather than the ordinary. Hutton (2005) uses
Garland‟s distinction between structuralist and individual accounts to explain the
difference. Structuralist accounts, dealing with how the system operates, are not based
on accurate information about crime and risk but on an account which expresses
anxieties about broad patterns of social disorder. In contrast, individualised accounts
elicit a distinctive response and with a case scenario, the respondent tries to find a just
solution for an individual case. These theories help explain why in our study, despite
improved levels of knowledge and their own recommended sentence being the same
or less severe than the actual sentence imposed, many respondents still tend to the
belief that, in general, sentences are too lenient.
79
Jury Sentencing Survey
Insights from the interviews
The advantage of our study is that we could examine this gap at three different stages:
first before the respondents knew of the sentence, secondly, after they knew of the
sentence and had more knowledge of sentencing patterns from the booklet, and finally
in the interviews. Stage 1 showed that there was a dichotomy between punitiveness as
measured by sentence choices in the individual case and abstract views as to
sentences in general. Stage 2 showed that, for sex and violent offences, the dichotomy
persisted between these measures of punitiveness after jurors were better informed.
The interview phase of our study provides insights into why some jurors were able to
bridge the perception gap and some were not. The following two case studies, drawn
from our interviews, illustrate these differences.
Alfa 1 was a juror on a „motor manslaughter‟ case. At Stage 1 she had selected
sentences for the two co-offenders which were very similar to the judge‟s sentences
but which were marginally more lenient in respect of the time which had to be served
before eligibility for parole. Her Questionnaire 1 responses indicated that she thought
sentences for violent and property offences were about right, but that sentences for
drug offences were much too lenient, and that sentences for sex offences were a little
too lenient. At Stage 2 her responses were that sentences for violent, drug and
property offences were about right but, in the case of sex offences were a little too
lenient. At the outset of her interview Alfa 1 said that after the trial and learning of the
sentence imposed by the judge, she no longer thought that in general sentences were
too light. Alfa 1 said that she had been quite traumatised by the trial experience. She
had not expected to feel sympathy for the offenders yet she had felt sympathy for
„both sides‟. She said:
I thought I was going to [say], „Yes, that person, they need to go to jail. That
was really bad.‟ But I thought I was going to be like that but I wasn‟t at all.
It was like, „This is terrible‟. This person‟s [the defendant] only 21. They‟ve
barely moved out of home. …
They‟re just normal people. [The offender‟s] mother did not raise her son to
kill someone else. And I‟d look at her and think how devastating it must be
for her to sit through this every day. She didn‟t raise her son to abuse
alcohol. Like she didn‟t want that when she had a son.
Alfa 1 saw real people in the courtroom and this made her more lenient towards the
offenders than she had expected. It also impacted on her general attitudes to
sentencing. Alpha 1 was clearly one of those who could extrapolate from her
increased knowledge and experience. At Stage 2 she answered all the crime and
sentencing knowledge questions accurately. She said:
So you generally – well, I know that I generally feel that sentencing is too
light. But sitting through this case every day and listening to complete
backgrounds… But then when I‟d done the survey before sentencing
actually happened and then it was – it [the sentence] was in the kind of area
that I thought it should be – is that – it‟s not – I don‟t think it is lenient. It‟s
just that you are privy to so much more information and there is so much
more to it than the sensationalist stuff you hear at six o‟clock.
And then later when discussing changes in her general views as to sentence, she said:
They‟re not lenient – I mean they‟re too lenient I would have said
beforehand.
80
Part 4: Discussion
In contrast, Charlie 2 was not an extrapolator but rather an „exception maker‟. His
views were particularly striking because he was on two trials and reacted in much the
same way to each. Despite the fact that he had not chosen a more severe sentence than
the judge and agreed with the sentence imposed, in Stage 2 and in the interview he
maintained his view that sentences for sex, violent, drug and property offences were
too lenient. Offender 1, who was on trial for assaulting his partner, had prior
convictions for assaulting women. Offender 2, who was on trial for cultivating a
controlled drug for sale, also had prior drug convictions. However, Charlie 2 saw
neither as a „real criminal‟. In relation to the defendant convicted of assaulting his
partner, he said:
He did not come across as a dangerous criminal. … I saw a human situation
that had gone off the rails for various reasons and no criminals really
involved, is what I saw.
Of the drug offender, he said:
But this fellow, he‟d just got into hard times and he could see a way out. He
wasn‟t – he wasn‟t similarly, he wasn‟t a bad sort of person, I didn‟t think.
He was just a smoker and he saw a way of – I mean it‟s obviously wrong
and the jury agreed that it was definitely wrong to be doing it but – oh look,
he was just a chancer, really.
And later he added:
So, in this case, you know, I‟ve almost got a bit of fondness for this bloke in
a kind of rough diamond sort of way.
His views of the offenders and the sentences in these two cases contrasted with his
general view that sentences were too lenient. In discussing his general views, Charlie
2 explained that he had a different picture in mind of the kinds of people who were
being let off too lightly for crimes of sexual violation and drug trafficking.
In the case of sex offenders:
I wasn‟t thinking about the trivial thing we had seen. [Offender 1 was
charged with sexually assaulting his partner but was acquitted on this count.]
I was thinking about the nightclub bloke who slips someone a drink and then
takes them up and there‟s a gang rape going on. To me that‟s absolutely
diabolical.
And in the case of drug offence sentences that he considered too light:
I probably wasn‟t thinking about this bloke who was an amateur. … So I‟m
thinking about the people that are in it to make huge amounts of money
quickly and with no thoughts of human misery or anything like that.
This explanation was illuminating. It resonated with suggestions made by other
researchers that members of the public who respond to polls tend to construct
stereotypical pictures of the worst kinds of offenders that reflect the images
disseminated in the media and popular culture of violent, ruthless, pathologically evil
predators who are „sick, mad or bad‟ (Roberts 1997: 113; Unnever and Cullen 2009).
Charlie 2, like many others interviewed, clearly recalled the violent and the extreme
rather than the ordinary (Diamond and Stalans 1989: 87). Others, like Echo 1, also
distinguished between the offender in her case, who had been convicted of
maintaining a sexual relationship with a young person, and those whom she thought
of as typical sex offenders:
81
Jury Sentencing Survey
[He] wasn‟t pure evil, that kind of, you know the way that you would usually
look at a paedophile or ... the ones that seek victims out and plan quite
nastily to do degrading or nasty or forceful things to them ... he was atypical
of what I was expecting.
Likewise, Foxtrot 1 was able to make an exception for her case:
I mean this was an odd case really. You know, a lot of us probably, I felt,
that it shouldn‟t have been in the court and yet, once it was in court, yes it
was an assault case but with a difference.
Mike 1 also made a distinction between the offender in her case, who had been
convicted of attempted murder, and the „criminals you see on TV, you see them as
repeat offenders.‟ Oscar 2 also contrasted the images from television with the „silly‟
offender in the dock who had „a poor family life:
[Y]ou see dreadful things on television don‟t you and you think, oh they‟re a
terrible type, but he didn‟t really fit that picture in my opinion.
X-Ray 1 said of the offender, who had been convicted of a sex crime, that he was
„certainly not a criminal in the true sense of the word‟:
[I]t‟s not like he was a – it‟s not like – some people are perpetual reoffenders and [you] probably can‟t ever help some of those people. They‟ve
obviously got a problem, you know. It probably is a really mental problem
some of these perpetual rapists or whatever. They‟ve obviously – whereas
this guy didn‟t come across like that at all. He was just an idiot.
We have called Charlie 2 an „exception maker‟ because he, like many other jurors
interviewed, made an exception for his two cases, which did not seem to involve
„real‟ crimes or „real‟ criminals and he sought to distinguish them from what he saw
as the more typical serious cases where judges were too lenient. The contrast between
these views and those of Alfa 1 showed that there is more to changing public opinion
than seeking to expose members of the public to more information about actual cases.
Both Alpha 1 and Charlie 2 recognised real people in their trials and sympathised
with them as individuals caught up in difficult situations. They were not in any way
excessively punitive towards them. However, Alfa 1‟s case involved a serious and
prolonged piece of dangerous driving that resulted in the gruesome death of an
innocent motorcyclist, whereas Charlie 2 saw two cases that did not fit in with his
picture of serious crime. The seriousness of Alfa 1‟s case did not give her the
opportunity that Charlie 2 had to make an exception for her experience even if she
had been inclined to do so.
These reactions suggest that seeing the offender as a real person and being given more
information about sentencing practices and processes is not the only key to changing
perceptions. Charlie 2 answered the crime and sentencing knowledge questions
correctly at Stage 2. He knew that his sentence was less than the judge‟s. He no
longer had misperceptions about crime and sentencing. He was not fearful about
crime (he said he felt very safe at home alone at night and fairly safe walking alone at
night). But still he thought that sentences were too lenient. Hutton‟s explanation for
the difference between structural and individual accounts does not entirely explain the
perception gap in Charlie 2‟s case. His „structural account‟ was not based on
misperceptions about crime or risk and he appeared to have no anxieties about social
disorder. Rather, his misperceptions lay in what saw as a real sex or drug crime as
reported by the media.
82
Part 4: Discussion
As well as exception-making, we saw examples of respondents who tried to find „a
just solution‟ for their own individual case (Hutton 2005) that was often more lenient
than the judge‟s sentence, but who at the same time maintained their general views
about sentencing leniency. So, jurors often adopted a practical problem solving
approach that was aimed at reforming individual offenders and encouraging them to
turn way from their criminal behaviour. Golf 1, for example, did not want to see the
offender in her case locked up, rather:
I just thought she had a bit of an insular outlook and, sort of, she needed to
get out a bit more. ... I just thought it would be good for her. I just thought
she was a bit self absorbed. ... She needed to get out a bit more and it‟d be
good for her to contribute a bit.
Juliett 1 also wanted to see the offender perform community service:
I don‟t think he saw outside his own four walls, really. I do believe that he‟d
gotten a bit wiser as he aged because the offence had happened a number of
years beforehand, but I still perceived a “poor me” kind of attitude from him
so that was part of the reason for the community service idea.
Oscar 1 thought that general sentencing levels for violent offenders were a little too
lenient; for property offenders, a little too lenient; for drug offences, much too lenient
and for sex offences, much too lenient. However, in the drugs case that he decided,
Oscar 1 wanted to see a wholly suspended sentence and community service for the 71
year old offender on the grounds that „it wouldn‟t be human‟ to send him to gaol.
5. Informed members of the public do not consider judges are as ‘out of
touch’ as populist public opinion suggests
Judges are often portrayed not only as being too lenient, but also as being out of
touch. Top-of-the-head responses from representative surveys in the UK show that a
high proportion of respondents state that judges are very out of touch and only 18 to
20 percent responded that they were in touch (Hough and Roberts 1998; MirleesBlack 2001: 5). Our participants, who sat through one or more trials in a courtroom,
had at least some first-hand knowledge of judges. Rather than responding that judges
are out of touch, more than two-thirds of respondents in Stage 1 said that they were in
touch (very or somewhat). And after the sentence was imposed and they had received
the judge‟s comments, the response improved to 82 percent. This result accords with
research which suggests that jurors have more confidence in the criminal justice
system than other members of the public (Maruna and King 2004: 12). This finding
also suggests that there may be some benefit in sending the judge‟s sentencing
comments to jurors because it may add to the improvement in public confidence in
criminal justice that jury service brings.
Other studies have found that perceptions that judges are out of touch are correlated
with perceptions of judicial leniency (Hough and Roberts 1998; Hough and Roberts
1999). Unsurprisingly, in this study, the more in touch judges were perceived to be,
the less likely sentences in general were said to be too lenient (see Table 44) and the
less likely participants were to want a more severe sentence than the judge in the case
they tried (Table 46). In fact, on all five measures of punitiveness (Stage 1
comparative sentence severity, Stage 2 appropriateness of sentence, Stage 2
comparative sentence variable, and jurors‟ views as to sentencing in general at Stage
1 and Stage 2) the more in touch judges were perceived to be, the more lenient were
83
Jury Sentencing Survey
the respondents. The relationship was strongest for general attitudes to sentence. It
follows that improving perceptions about judicial remoteness can impact both on
perceptions of judicial leniency and on public punitiveness.
The interviews provided the opportunity to further explore the responses to this
question. Many jurors had responded that judges were „somewhat in touch‟. Some
respondents thought that the judge‟s role may have desensitised them from realising
the seriousness of crime, especially when compared with the community‟s view.
Hotel 1 explained that it was:
[B]ecause they‟re involved in so many [trials] I think they do become a little
bit hard and blasé about it. … Nothing surprises or shocks them any more so
it becomes a bit of a matter of course through the day.
Discussion of this question led some respondents to question what is meant by „public
opinion‟, to wonder whether there was such a thing, and in some cases, participants
who had newly become aware of the misperceptions about crime and sentencing that
so many people have, were led to question the value of judges taking public opinion
into account. Zulu 1 said that his view changed after reading the book:
So, it‟s a bit hard to reflect public opinion when public opinion‟s not
informed, as mine wasn‟t before reading the book.
India 1, who thought that judges were very in touch with public opinion, was
nevertheless very sceptical about judges taking public opinion into account, because it
is not informed:
What I meant by that was sometimes public opinion is really swayed by
sensational nonsense in newspapers. So we are swayed, the public can be
swayed and then, if we are swayed, if judges then should reflect, then I think
that‟s really dangerous.... They shouldn‟t because, you know, if huge
percentage of the public are watching Today Tonight and A Current Affair,
then God help us if the judges then need to follow their opinions because it‟s
just for ratings. It‟s just sensationalised....
Other respondents were driven by their jury experience to question whether they
themselves were in touch with public opinion and to defer to the wider experience
that judges have. Tango 1 thought that judges were „somewhat out of touch‟ in
Questionnaire 1 and then, in Questionnaire 2, he moderated his view to „somewhat in
touch‟ but – although he maintained that, because of their isolated role, judges by
necessity have got to be out of touch – he also doubted whether he, as a member of
the public, was any better placed:
[I]f the courts are way out of kilter from what the public thinks the
punishment should be, the people have contempt for the process. ... but then
again, I was quite surprised when I read these figures, so it shows how out of
touch I am with what the courts do.
Alfa 2 said that „public opinion is largely over rated‟ and wondered whether the views
that are „whipped up by the media‟ were truly reflective of what the public actually
thinks. Consequently, Alfa 2 thought that, because of their experience, judges were
better equipped than ordinary members of the public to withstand media
manipulation. Golf 1 strongly agreed that judges should reflect public opinion and
although she initially wondered whether, because of their age, judges might have been
only somewhat in touch with public opinion, said, on reflection:
84
Part 4: Discussion
But then again, I‟m sure they‟re very exposed to all elements of society, so
maybe they‟re more in touch than I am.
Other interviewees made a distinction between judges in general and their judge.
While in general they thought that many judges were somewhat out of touch, because
of their privileged position in society, their stage of life, and the relative narrowness
of their circle of acquaintances etc, jurors nevertheless felt that the judge in their trial
did a very good job. Oscar 1 thought that judges in general were too lenient and were
somewhat out of touch with public opinion, but maintained that his judge „was
certainly in the ballpark as to what he was saying.‟ Papa 1 found her judge to „be very
fair‟ and „not so austere as they used to be‟ and Papa 2, who thought that judges were
affected by „the ivory tower type thing‟ and were „not in the real world‟ nevertheless
thought that the judge „certainly seemed to be in full grasp of exactly what happened.‟
In the first questionnaire, X-Ray 1 thought that judges were „very in touch‟, but by the
second, he had changed his view to only „somewhat‟ in touch. However, he
maintained that:
Look, I must confess I thought that [the judge] was terrific. Very balanced
and fair so it‟s hard to say because he‟s the only judge that I‟ve been in front
of.
This is a further demonstration of the „perception gap‟ and the difference between
general perceptions of judges as an anonymous group and responses to individual
judges and their personal performance during the trial.
The issue of judicial remoteness is obviously linked with the issue of the
responsiveness of judges to public opinion. In Stage 2, 65 percent of jurors thought
judges should reflect public opinion when sentencing. This was similar to the results
of the 2003 and 2007 Australian Survey of Social Attitudes (AuSSA), although juror
respondents were less likely to strongly agree with this statement than AuSSA 2003
respondents (Indermaur and Roberts 2005: 153; Roberts and Indermaur 2009: 20).
Those who agreed judges should reflect public opinion were more likely, after
knowing the sentence imposed by the judge and receiving the information booklet, to
have:

preferred a more severe sentence;

said sentences in general were too lenient;

said judges were out of touch.
The interviews provided participants more opportunity to reflect on this question and
some qualified or altered their views as a result. For example, Charlie 2, a juror who
at Stage 2 had agreed that judges should reflect public opinion, later said:
I think they need to be aware of public opinion. So therefore I might
question the word „reflect‟. I think „be aware of‟ might be better. … Public
opinion‟s often quite dodgy I think.
Another juror, Victor 1, said in the survey that she thought that judges were somewhat
in touch, but explained in the interview that she did not necessarily think that this was
a bad thing. She did not want judges to sentence „at the whim of popular opinion.‟ So,
this juror‟s assessment that judges were not completely in touch with public opinion
was not meant as a criticism:
85
Jury Sentencing Survey
I don‟t think judges should be responding a hundred percent to popular opinion
because they do have a set of criteria that they‟re working to. However, if overall
there‟s not a reasonable degree of overlap of those two then there‟s a serious
problem in our society. So, in the big picture, the scheme of things, judge‟s
sentences need to be broadly aligned with community views, but not with
regard to a specific case.
So, the mere fact that some members of the public suggest in their survey responses
that judges are not in touch with public opinion, cannot necessarily be interpreted as a
criticism of judicial sentences. At least some of those responses can be explained by
the fact that these members of the public have confidence in the judiciary and trust
them to sentence fairly – and distrust the views of their fellow citizens.
6. Jurors, in common with other members of the public, are poorly informed
about crime and sentencing
Our study showed that jurors, like other members of the public, are poorly informed
about crime and sentencing. The misperceptions at Stage 1 included:

that crime levels are rising (63%);

that burglary and motor vehicle theft are rising (57%);

overestimating the proportion of crime that involves violence (75%);

underestimating the imprisonment rate for burglary (80%) and rape (71%);
and

overestimating the risk of being a victim of burglary (67%), motor vehicle
theft (62%), assault (45%) and robbery (69%).
Given that electronic and print media are the public‟s main source of news (Denemark
2005: 223; Roberts and Indermaur 2009: 9) and that our jurors‟ major sources of
information about crime and sentencing were also newspapers, television and radio,
their misperceptions revealed at Stage 1 are not surprising. The media focuses on the
violent crime and sensational crime stories rather than ordinary crime, and on lenient
sentences rather than harsh sentences, thereby leading the public to overestimate the
amount of crime that is violent and to underestimate the severity of sentencing. Once
they have been given the opportunity to reflect on their knowledge – and the sources
of their knowledge – our jurors readily admitted in the interviews that their opinions
were not based on any detailed sources of information.
Bravo 1: Well, I have no idea really about these things.
Delta 1: It‟s just a perception. It‟s not based on any facts. It‟s a feeling.
Lima 1: [N]o particular reports –I‟m just generalising.
Hotel 2: I have to preface my remarks by saying that I‟m really no expert on
this and these are really just impressions.
Tango 2: Pure, uninformed speculation I guess ... I‟m thinking of the great,
the lines, the constant themes that we‟re fed by the media, really, because I
have no personal experience in any of those. .... Yes, because if you ask me
the last rape case that was in front of the court here, what the sentence was;
no idea. I‟m not following any specifics. I have no data to back up my
opinion.
86
Part 4: Discussion
X-Ray 2: I‟m not really up on what‟s actually happening, but just a general
sense of, you know, my, my gut feeling....
Lima 2 reflected on the media sources of information and contrasted them with the
Booklet:
I suppose it laid out some facts, more than anything else. It gave me a
probably truer indication of, of sentencing and crime and statistics of crime
and, and trends or otherwise. Because it‟s easy to fixate on the last thing that
was, that was high in the media and maybe two of those incidents occurred
in one week and you suddenly think, you know we‟re at a period of, of a lot
of home invasions, when in actual fact they‟re probably down ten percent on
previous years so, to actually see the, the figures and see the trends was
really good because the perception, and it‟s not pushed by the media, but
there‟s just, it‟s at the forefront of your ... consciousness at the time.
This suggests that the combination of jury service and the provision of better
information can lead some members of the public to reassess their views. The issue of
whether this can lead to a drop in punitiveness is discussed below.
7. The better informed (and the least fearful) are the least punitive
Our study confirms the findings of previous studies (Gelb 2006: 15) that crime
misperceptions influence perceptions of leniency in sentencing. People who know
more about crime are less punitive (Roberts and Indermaur 2007: 61-62; Roberts and
Indermaur 2009: 19). A belief in rising crime, for example, is often coupled with a
belief that lenient sentencing is the major cause, thus perpetuating the perception that
sentences are too lenient (Roberts and Hough 2005: 48). Our results showed that
perceptions of lenient sentencing at Stage 1 were associated with the beliefs that
crime had increased, with overestimates of the proportion of crime that is violent,
with under-estimating the imprisonment rate for rape, and overestimating the risk of
victimisation. Uniform 1, for example, who thought that judges were very much too
lenient for all types of offences, was also misinformed about sentencing levels,
saying: „I mean, like, you go and murder someone and you what? You get two years?‟
Additionally, those who were more fearful walking home after dark or being home
alone at night were significantly more likely than those who were less fearful, to
perceive sentences as too lenient. This also conforms with the findings of previous
studies. Those who were more fearful were also more likely to have misperceptions
about crime and sentencing suggesting that removing misperceptions could have a
positive impact on reducing fear as well as on punitiveness.
8. Information improved knowledge about crime and sentencing
After respondents had received the crime and information booklet, their knowledge
improved. However:

38 percent of respondents still said that recorded crime rates had increased;

37 percent still over-estimated the proportion of crime that involves violence;

35 percent still under-estimated the imprisonment rate for convicted rapists;
and
87
Jury Sentencing Survey

about half still over-estimated the risk of being a victim of burglary or motor
vehicle theft.
This was despite the fact that 62 percent had read the booklet in full and an additional
23 percent had read at least some sections. Moreover, respondents indicated the
booklet was easy to understand. Sceptical reactions to declining crime rates have been
consistently reported from Canada, the US, the UK and Australia. „These statistics
can‟t be right!‟ is a common response. Moreover, accepting the view that crime rates
are decreasing can be seen by some respondents as being complacent and
unconcerned with current levels of crime (Roberts et al 2003: 13). In the interviews
we made the conscious decision not to test participants‟ knowledge of crime and
sentencing because we did not want make them feel uncomfortable. However,
remarks of some of the respondents made it clear that no faith was placed in official
statistics.
Hotel 1, for example, still responded that recorded crime had increased a lot in
Questionnaire 2 even though he had read the booklet in full. In his interview he gave
examples of witnessing young people shoplifting and rudely brushing of security staff
to demonstrate his belief that „crime is starting at a younger age, 12, 13, 14 and
they‟re not allowed to be touched.‟ His personal experience supplied the proof that
crime rates were rising. The view that „The statistics can‟t be right‟ explains his
response. Another respondent who had read the booklet in full and responded that
crime had increased a lot, added the comment on the questionnaire that it „appears to
have increased the media reports more crime‟. Even if respondents‟ knowledge
improved so they could correctly answer the question in Stage 2 and respond that
recorded crime had decreased, they could still be sceptical as to what this means in
terms of crimes actually committed. For example, one respondent who responded
recorded crime had decreased added the note: „I know people don‟t report crime
because cops don‟t come.‟ These responses probably reflect a deeper scepticism
among some sections of society of whether recorded crime is a good reflection of
actual crime.
9. In some respects punitiveness dropped after receiving more information
but respondents are not always consistently punitive
After jurors had received the judge‟s sentencing remarks and the Crime and
Sentencing booklet, punitiveness dropped in the sense that fewer respondents wanted
a more severe sentence than the judge. At Stage 1, 45 percent of jurors who went on
to participate in Stage 2 selected a more severe sentence than the judge. But at Stage 2
this had dropped to 37 percent. This drop was clearly apparent for violent offences
with 49 percent selecting a more severe sentence at Stage 1 and 35 percent doing so at
Stage 2. Similarly 50 percent said that the sentences for violent offences at Stage 2
were very appropriate. Of those who had said that the sentence for the violent offence
was very appropriate, 44 percent had selected a more severe sentence. However, for
sex offences there was no such drop in punitiveness with 45 percent selecting a more
severe sentence at Stage 1 and 46 percent preferring a more severe sentence at Stage
2. (Only 36 percent said the sentence was very appropriate, of whom only 19 percent
had chosen a more severe sentence). (See Figs 9, 11 and 13).
Whilst the views of respondents had moderated for all offence categories except sex
offences, there was a difference, as noted above, between those who had selected a
88
Part 4: Discussion
more lenient sentence and those who had selected a harsher sentence. Those selecting
a more severe sentence were less likely to defer to the judge‟s sentence than those
who selected a more lenient sentence. This accords with Lovegrove‟s (2007) findings.
As also discussed above, it appears to suggest that while the harsh may moderate their
views, their views are less malleable than those who are more lenient. This was
confirmed by an analysis of preferred sentence at Stage 2. Those who had chosen a
more severe sentence in Stage 1 were less likely to endorse the judge‟s sentence in
Stage 2. Because of this it cannot be categorically stated that in general respondents
had become more lenient. Some (19%) had become more lenient, but because others
endorsed the judge‟s more severe sentence, they had, in this sense, become more
punitive at Stage 2 (43%).
Punitiveness as indicated by abstract views (based on the question about whether
sentences are too tough, about right, or too lenient) decreased after the receipt of the
judge‟s sentence and the information booklet. Across all four offence types,
respondents were less likely to say sentences are too lenient (see Table 29 and Table
30). Of course it is not possible to say that this was because of improved knowledge
only (Chapman et al 2002, 50). The change in attitude may be because of the process
of engaging in the study, the „Hawthorne effect‟ or other unknown reasons. As
discussed above, another demonstration of decreased punitiveness is the change in
response to the question about whether judges are in touch. Figure 22 shows that „out
of touch‟ responses decreased and respondents were more likely to say judges were
„very in touch‟ in Stage 2 after receiving the sentencing comments and booklet.
The interviews were able to explain some of these shifts in apparent punitiveness.
Some jurors, who had imposed a much more severe sentence than the judge at Stage
1, appeared on this measure to be highly punitive. However, once they were asked to
elaborate upon the reasons for their choice, it became apparent that they were not
necessarily motivated by punitive attitudes at all. India 1, who had suggested a
sentence of seven years in a case where a father had sexually abused his daughter,
explained that she had selected the number as a symbolic statement. The father had
abused his daughter over a period of seven years and so India 1 chose that penalty to
match the period of abuse, not really thinking that it should actually be imposed. In
fact, India 1, who had been given a glimpse of the prison system by sitting through an
inquest into deaths in custody, was very much opposed to imprisoning offenders
merely for the sake of punishing them:
Look, I hate the idea of anyone going to jail. ... I really hate it. So having
seen the inside of what the jail is like and listened to – you know, it is a
horrendous place.
India 1 strongly expressed support for rehabilitation and for restorative practices:
Jail shouldn‟t be the punishment. ... Actually, if I thought that when they
were incarcerated they were going to get an education, learn a trade, steered
in a different direction ... But as it is, I don‟t believe in – I don‟t believe in
long sentences.
Sierra 2, who had also imposed a more severe sentence than the judge, was motivated
by a strong view that the offender needed rehabilitation and that a longer period of
detention would allow for a greater degree of rehabilitation. Sierra 2 said that the
offenders needed not just „prison‟ but some „corrective services‟ so that „there‟s a
little glimmer of hope for the criminals that go in.‟
89
Jury Sentencing Survey
You can‟t just put them in for two years. They‟re going to go through a
process that takes, probably, six to eight months just assimilating them into
the system. ... Therefore I say ... that he ... would not have long enough to
(rehabilitate) – not that you can probably redress it to a great degree, but you
might nurture and help something in him further.
So, some jurors who appeared to be harsh at Stage 1, were in fact motivated by more
positive pro-social emotions. Others, like India 2, who appeared at Stage 1 to be
rather more lenient than the judge but who became harsher at Stage 2, had a different
explanation for the change. India 2 explained that her initial response at Stage 1 had
been influenced at the time by the fact that she had had „terrible trouble‟ coming to
her decision and had experienced an upsurge of emotion when she had to give her
verdict. She also sympathised with the offender‟s family and felt very sorry for the
defendant, who seemed like „just an average Joe‟. However, after the emotions had
subsided and she had taken some time to think about the sentence, she decided that
the judge‟s harsher sentence was the more appropriate one.
I‟m the sort of person that, if someone tells me something, I can often sleep
on it, come back the next day and I know clear in my mind. I just need that
little bit of extra time. Yes, I just didn‟t feel as though I had enough time to
make that decision. Where now, I sort of – I‟m happy with the result and
that, but at the time I sort of walked away a bit unsure, you know. But now
I‟ve had the time to think about it, I‟m happy.
Another juror, November 1, also reported that she had been influenced in her Stage 1
sentencing decision by her emotions and by the pressure that she felt from other jurors
who had expressed harshly punitive views about the offender. However, by contrast
with India 2, who became harsher with the passage of time, November 1 became
more lenient.
You go home; you‟re full of emotion and all that and then I let them, I guess,
tell me things which I didn‟t believe in my mind. So I guess when I then got
them out of my mind and thought about it I thought, “No, well...”
These examples show once again that attempts to measure public opinion must take
into account the fact that it is nuanced and complex – and that a single respondent‟s
views may appear to shift markedly over a fairly short time, depending on the kinds
of questions that are asked. Many of our respondents cannot be described as being
unequivocally „harsh‟ or unambiguously „lenient‟ in their attitudes to criminal
sentencing and so, while an individual juror‟s views could readily be understood once
we had examined the detailed responses given in the interviews, the statistics when
considered on their own, can, in many cases, be ambiguous or misleading.
In summary, on a number of measures, punitiveness dropped with improved
information. Respondents were less likely to say sentences in general were too
lenient; fewer respondents wanted a more severe sentence than the judge in the case
they had heard, and respondents were more likely to say judges were in touch. But
because many of those who had selected a more lenient sentence than the judge at
Stage 1, then agreed with the judge‟s sentence at Stage 2, we have to conclude that
those participants had become more punitive with more information – at least in that
respect.
90
Part 4: Discussion
10. Providing more information and improving knowledge is not a panacea
Our study shows that, overall, improving information about crime and sentencing
reduces punitiveness in sentencing attitudes but not in a uniform way. The provision
of information is not always enough to change attitudes. This is because attitude
formation is a complex process and is not simply a function of lack of knowledge. For
some respondents the belief that sentences are too lenient is firmly entrenched and is
not shaken by the knowledge that sentencing practice is tougher than they thought, or
that in a real case in which they determined guilt they would have selected a more
lenient sentence than the judge did. This study suggests that there is more to be learnt
from exploring the reasons why some participants could not „jump the perception
gap‟. One reason we propose from our results in this study is that such a failure is not
a lack of knowledge of crime or sentencing trends but the misperception that the
stereotypical offender is the typical. This, however, does not make the views of such
people less real or relevant. Even when fully informed, they may consider sentences
for the type of offender they had in mind are too lenient. This can only be properly
tested by further research. But it does mean that surveys of public opinion, which
suggest that the general sentencing levels for a particular type of offence are too
lenient, cannot be taken at face value.
11. Jurors can act as conduits of information to the wider community
Results from the surveys and the interviews suggest that there would be benefits in
both improving the information given to jurors and in providing them with the
sentencing comments made by the judge in their case (or in the alternative,
knowledge about how to access those comments). Ninety seven percent of the sample
thought jurors would be interested in receiving the Crime and Sentencing booklet.
Others thought it should be more widely available. Zulu 1 suggested:
That‟s something that I think could probably even get some circulation
within schools or later years of school life.
Victor 1 also thought that:
Overall I thought it was an excellent book and I would like to find that sort
of book in my doctor‟s waiting room. … If these sorts of things were
available in the courts I think it would be helpful.
As reported above, about half of the juror participants in the study said that knowing
the judge‟s reasons for sentence affected their view of the appropriateness of the
sentence „a lot‟. Almost all participants (98%) also thought that jurors would be
interested in knowing how to access the judge‟s sentencing comments. Jurors tend to
become very engaged in their case and this stimulates their receptiveness to
knowledge about crime and sentencing information as the comments of November 1
and Whiskey 2 show:
November 1: When you watch TV and read the papers they make it out to
sound a lot worse. … It‟s interesting because I‟ve gone through a few of [the
judge‟s] other cases trying to determine if I agreed with the judge.
Whisky 2: I actually looked back at similar cases that were on at the
Supreme Court at the same time as well, just out of interest.
The results also suggest that jurors can act as conduits of information to the general
public. A third of the respondents discussed the information about sentencing in the
91
Jury Sentencing Survey
booklet with colleagues and friends and more than two thirds discussed the sentence.
The following comments from interviews illustrate how jurors can act in this way:
Delta 1: I worked with a very flippant group of people and I walked into
work and they said, “Oh God, you let him off. You didn‟t hang him.” But
when you actually talk to them they were all satisfied with the result.
Foxtrot 1: My father is in Melbourne and I told him I had been on a jury. …
He said, “Oh that was one of those sensationalist current affairs stories.” …
He was very unsympathetic about both people [offender and victim] and I
said, “Dad, you had to be there”.
India 2: As soon as I told people what it was about afterwards, it was a 49
year-old and a 17 year-old. “He should have been castrated,” they said. … I
got the [comments] up on the Internet and said, “Look, read this”.
B. ANSWERS TO THE RESEARCH QUESTIONS
The results of the study suggest the following responses to the six research questions
formulated for the application.
1. How can juries be utilised as a source of public opinion about
sentencing?
A good proportion of jurors are willing to participate in a study that explores their
views about sentencing. Many of them are willing to stay after the verdict and listen
to the sentencing submissions and then complete a questionnaire, notwithstanding the
fact that they also report feeling the burden of jury duty. Once they have participated
in this way, many are willing to continue to participate and to read the information
provided and complete another questionnaire. Many agreed to be interviewed and
some even welcomed the opportunity.
Both the willingness of jurors to participate in the sentencing survey and the fact that
they appear to be fairly representative of the Tasmanian population, suggests that
jurors are a good source of public opinion. Using jurors also has the advantage that
the jury is a well-respected institution and so their views are likely to be given some
weight. Another advantage is that jurors are not merely well informed about the facts
of the offence, they have also had the opportunity to observe the defendant through
the course of the trial. This gives them a sense of the offender as a real person,
something that became very apparent in the course of the interviews. These features
give the jury survey method an advantage over studies that use sentencing vignettes
which, even if they use real cases, cannot impart this personal element.
The second advantage this method has over surveys or focus groups that use
vignettes, is the element of responsibility for decision-making that the jurors have
shouldered. This responsibility means that they have a real sense of the burden of
decision-making that the judges normally experience when they must sentence the
guilty offender. In the literature these two aspects of decision-making have been
identified as important.
In addition to the provision of information and the assumption of responsibility, the
third aspect of deliberation, is (partly) achieved by the opportunity that is given to
92
Part 4: Discussion
participants to reflect on their decision after reading the sentencing remarks and the
information in the Crime and Sentencing booklet. The Stage 3 interviews give further
opportunities for discussion and deliberation, and are useful to flesh out and explain
the quantitative results from the first two stages.
2. How receptive are jurors to learning about crime trends and sentencing?
The answer to this question is that jurors are quite interested in learning more on both
these areas. Over 60 percent of Stage 2 participants read the Crime and Sentencing
booklet in full. A further 23 percent read some sections in full. Most found the
information easy to follow and interesting. While jurors‟ knowledge of crime and
sentencing matters improved considerably as a result of reading the booklet provided,
many still retained common misperceptions about crime trends in particular. There
was some evidence from the interviews that this was because of a general scepticism
and mistrust of official crime statistics.
3. To what extent are jurors (as newly informed members of the public)
satisfied with the sentence imposed by the judge?
Our study found that ninety percent of jurors were satisfied with the sentence and half
of these were very satisfied with it. Jurors were less satisfied with sex and drug
offence sentences and almost half of jurors on sex and drug offence trials would have
preferred a more severe sentence. This result indicates that when jurors are aware of
the complexities of a case and the judge‟s reasons for imposing sentence, the majority
will then agree with the sentence imposed by the judge.
4. What kind of information affects public satisfaction with sentencing?
Our study aimed to test three specific types of information for their impact on
sentencing satisfaction:

Listening to the sentencing submissions.

Knowledge of crime trends.

Information about sentencing law and sentencing patterns.
The results suggest knowledge of crime trends and sentencing information can
operate to affect sentencing attitudes but our methodology had limitations. For
example, we could not fully assess the impact of the above factors on satisfaction with
sentencing because participants were only asked about the sentence at Stage 2. By this
stage all had heard or read the sentencing submissions, and had read the booklet with
the information about crime trends and sentencing patterns. It was not feasible to split
the sample into those who heard (or received) the sentencing submissions and
received the booklet and those who did not. However, Stage 2 participants were asked
about whether the sentencing information in the booklet was useful when they formed
their judgement of the appropriateness of the sentence. Only 10 percent did not find it
useful and 37 percent found it very useful. They were also asked about the impact of
the judge‟s sentencing comments and the information on crime levels on their
judgement of the appropriateness of the sentence. While most respondents said that
93
Jury Sentencing Survey
knowledge of crime levels and trends did not have a lot of impact on their judgement,
a half said it affected it a little. However, for half (51%) the sentencing comments
affected their judgement of the sentence a lot and another 41 percent said it had some
effect. This suggests that there are advantages in making sentencing comments
publicly available.
5. What variables affect jurors’ satisfaction with sentence?
Given that 90 percent of participants were satisfied with the judge‟s sentence and less
than two percent thought it was very inappropriate, there seemed little point in
seeking to determine if variables relating to demographic, offender or victim
characteristics affected satisfaction with sentence. However, differences between
offence types were explored. Jurors were less satisfied with sex and drug offence
sentences, with smaller percentages of very appropriate sentences (36% and 35%)
compared with violent and drug offences (50% and 57%). Moreover, almost half of
participating jurors on sex and drug offence trials would have preferred a more severe
sentence compared with 35 percent and 28 percent for violence and property.
Respondents were most satisfied with property offence sentences with just 28 percent
wanting a more severe sentence, and nine percent a less severe sentence. In fact,
comparing the sentence choice of respondents with the judge‟s sentence showed that
68 percent of property offence participating jurors selected a sentence that was more
lenient than the judge.
6. To what extent do the views of jurors as members of the public coincide
or differ from those of the judge as expressed in the sentencing
comments?
The research methodology proposed to explore this question in three ways: by
analysing the answers to open-ended questions asking whether there was anything the
participant particularly agreed or disagreed with; by comparing the judge‟s main
sentencing goal with the goal identified by the respondent as most important; and by
comparing the aggravating and mitigating factors identified by jurors with those
identified by the judge. The methodology proved to be not appropriate to answer this
research question. First, the response to the open-ended questions was poor. Only 17
percent indicated disapproval of anything in the sentencing comments and the
responses were difficult to categorise. Agreement with the comments tended to be
general rather than related to specific matters. Comparing the judge‟s aggravating and
mitigating factors with the juror‟s proved difficult. As explained in Part 3, while
analysis of the sentencing comments was capable of revealing the factors mentioned
by the judge as relevant, to determine to what extent the factor was very important,
quite important, not very important or unimportant proved too subjective for
comparison with a juror‟s opinion in the same case. Moreover, in the interviews it
became apparent that the „did not arise‟ column in Questionnaire 2 (see Question A7)
could be misunderstood. Alfa 1, for example, ticked „did not arise‟ in relation to prior
convictions because she read this as meaning „did not arise in the trial‟ rather that at
sentencing. Further refinements to the methodology are needed to analyse differences
between judges and jurors as to sentencing goals and aggravating and mitigating
factors. For example, rather than relying on interpreting the judge‟s sentencing
remarks, the judge needs to be asked the same question as the juror.
94
Part 4: Discussion
C. CONCLUSIONS AND POLICY IMPLICATIONS
The results of this study demonstrate that:

Jurors can be used as a source of informed public opinion;

Jurors can be used as conduits of information to better educate the general
public; and

Jury surveys can be an effective strategy to counter apparent public
punitiveness.
1. Using jurors as a source of informed public opinion
Asking jurors about sentencing is a useful approach both to measuring public opinion
and to understanding it. Jurors are willing to participate in reasonable numbers in such
a survey and in this study at least, they were reasonably representative of the general
population. This method has a number of advantages over representative surveys,
focus groups and deliberative polls, such as cost, engagement and legitimacy.
Because it can cover numerous examples of different offence types it provides a better
method of investigating differences in public opinion in relation to sentences for
categories of offence, than the traditional vignette methodology.
2. Jurors can be used as conduits of information to better educate the
general public
Jurors in this study were willing to read the sentencing comments and a booklet about
crime and sentencing. The fact that this not only improved their knowledge of such
matters and had an impact on their views about the appropriateness of the sentence,
but also that they had discussed the sentence in their case with others, demonstrates
that jurors can be used as a means of better educating the public about crime and
sentencing. However, there are limitations on the effectiveness of any strategy based
on providing better information to jurors in order to improve public confidence and
change attitudes. First, jury trials are rare and many members of the public do not
have the opportunity (or burden) of sitting on a jury. So, while a significant
proportion (68%) of jurors discussed the sentence with friends, the impact that this
might have on the wider community would not be a large one. Secondly, our results
revealed a „dichotomy‟ or „perception gap‟ between responses to an abstract question
about sentencing leniency and their responses to an individual case. This has
implications for the effectiveness of information as a means of attitude change. Some
respondents were shown to be unable to jump the gap from their own experience to
modify their perception of wider sentencing trends. These jurors would not
necessarily be effective ambassadors to the wider public.
3. The jury survey can be an effective strategy to counter public
punitiveness
The key finding of this study is that, informed members of the public overwhelmingly
approve of the sentences given by our judges. Based on the findings from 138 trials,
jurors who have judged the defendant guilty are more likely to select a more lenient
95
Jury Sentencing Survey
sentence than a harsher sentence than the judge. Moreover, when they are informed of
the sentence, they are highly likely to endorse it. The fact that this is the judgement of
jurors makes it a strong endorsement of judicial sentencing. It is an important finding
which should be heeded by politicians and policy makers. It suggests strongly that
jury surveys can help counter the „comedy of errors‟ – the situation in which policy
and practice is not based upon a proper understanding of public opinion and public
opinion is not based on a proper understanding of policy and practice (Allen 2002: 6).
Finally, in addition to the suggestion that jury sentencing surveys should be added to
the suite of methodologies used for measuring public opinion, the results of this study
suggest that there are advantages to providing better information to all jurors in the
form of a booklet about crime and sentencing after their deliberations. In cases where
juries have returned a guilty verdict, jurors should always be invited to stay and listen
(or to return and listen) to the sentencing submissions. And after the sentence is
imposed, jurors should be sent a copy of the sentencing remarks. In jurisdictions
where this is possible, they should also be informed about how to access sentencing
remarks from the court‟s website.
The jurors participating in this study were impressed with the judges who were
presiding over their trials and they felt more confident in the criminal justice system
as a result of their jury experience. This phenomenon has been remarked upon before
(Maruna and King 2004: 12), but our jury members, who stayed behind to hear the
sentencing submissions gained an extra measure of confidence, not only in the
criminal justice system itself, but also in their own verdict in the case and in the
judge‟s sentence as well. India 2, a juror who did not believe in the usefulness or
reliability of public opinion, explained that when she went into the jury she initially
thought: „Oh this is going to be difficult.‟ However, her experience was very positive
and she was reassured by having stayed behind to listen to the aftermath once the
verdict had been given.
It wasn‟t at all difficult. It was actually – I came away from it thinking,
“Yes, the jury system is excellent.” Because you do get perspectives that you
wouldn‟t – even if it was a jury of ten I don‟t think it would work as well as
the twelve … Having participated and gone through the whole process and
simply having stayed … really helped because I got to see the victim‟s
reaction … it was enough that she was believed. It seemed like she was
(believed) – and then I felt okay then. And, hearing the Judge‟s reaction to
the lawyers‟ submissions for sentencing I thought, “No, it will be – it is
going to be fairly dealt with here.”
Our findings suggest jury retrials should not be phased out. Rather, the trend for
increasing the jurisdiction of lower courts at the expense of jury trials should be
reversed. And the public should be encouraged to participate in jury service by
improving compensation and conditions.
96
REFERENCES
Aas KF 2005. Sentencing in the age of information: from Faust to Macintosh. UK:
Glasshouse Press
Allen R 2002. What does the public think about sentencing? Criminal Justice Matters
49: 6-41
Australian Bureau of Statistics 2006. 2006 Census of Population and Housing,
Tasmania (State) Tables Canberra: ABS
Australian Bureau of Statistics. Population by Age and Sex. ABS cat. No. 3210.0
Australian Bureau of Statistics 2005. Crime and Safety Australia Canberra: ABS
Australian Law Reform Commission 1992. Multiculturalism and the law. Report no.
57. Sydney: ALRC
Bartels LM 2008. Sword or feather?: The use and utility of suspended sentences in
Tasmania. PhD Thesis, University of Tasmania
Beale SS 2006. The news media‟s influence on criminal justice policy: How marketdriven news promotes punitiveness. William and Mary Law Review 48(2): 397-480
Benesh S 2006. Understanding public confidence in American courts. The Journal of
Politics 68(3): 697-707
Bloustein G & Israel M 2006. Crime and the media, in Goldsmith A, Israel M & Daly
K (eds), Crime and Justice: A Guide to Criminology, 3rd ed. Sydney: Lawbook Co
Bottoms A 1995. The philosophy and politics of punishment and sentencing, in
Clarkson C & R Morgan (eds), The Politics of Sentencing Reform, Oxford: Clarendon
Press: 17-50
Bottoms A 2004. Empirical research relevant to sentencing frameworks, in Bottoms
A, Rex S & Robinson G (eds), Alternatives to Prison: Options for an insecure society,
Cullompton: Willan Publishing: 59-82
Casey S & Mohr P 2005. Law-and-Order politics, public-opinion polls and the media.
Psychiatry, Psychology and Law 12(1): 141-151
Cashmore J & Trimboli L 2006. Child sexual assault trials: A survey of juror
perceptions. Crime and Justice Bulletin 102
Chapman B, Mirrlees-Black C & Brown C 2002. Improving public attitudes to the
Criminal Justice System: The impact of information. Home Office Research Study
245 London: Home Office
Chesterman M, Chan J & Hampton S 2000. Managing prejudicial publicity: An
empirical study of criminal trials in New South Wales. Law and Justice Foundation of
New South Wales
97
Jury Sentencing Survey
Connelly M, Wagner K & Jones G 2001. The public and correctional options in
Maryland: The results of a ‘deliberative focus group. State Commission on Criminal
Sentencing Policy
De Keijser JW, van Koppen PJ & Elffers H 2007. Bridging the gap between judges
and the public? A multi-method study. Journal of Experimental Criminology 3: 131161
Denemark D 2005. Mass media and media power in Australia, in Wilson S, Meagher
G, Gibson R, Denemark D & Wester M (eds), Australian social attitudes: The first
report, Sydney: UNSW Press: 220-239
Department of Police and Emergency Management 2009. Annual report 2008-09.
Hobart
Department of Police and Public Safety 2004. Annual report 2003-04. Hobart
Diamond S & Stalans L 1989. The myth of judicial leniency in sentencing.
Behavioural Sciences and the Law 7(1): 73-89
Doob A & Roberts J 1983. Sentencing: An analysis of the public’s view of sentencing,
Ottawa. Department of Justice Canada
Feather NT, Boeckmann RJ & McKee IR 2001. Jail sentence, community service, or
compensation?: Predicting reactions to a serious corporate offence. Australian
Journal of Psychology 53: 92-102
Fordham J 2009. Juror intimidation? An investigation into the prevalence and nature
of juror intimidation in Western Australia. Jury Research Unit, Centre for Forensic
Science, University of Western Australia
http://www.department.dotag.wa.gov.au/_files/juror_intimidation.pdf
Forsterlee L, Fox GB, Forsterlee R & Ho R 2004. The effects of victim impact
statement and gender on juror information processing in a criminal trial: Does the
punishment fit the crime? Australian Psychologist 39: 57-67
Freiberg A 2001. Affective vs. effective justice: Instrumentalism and emotionalism in
criminal justice. Punishment and Society 3(2): 265-278
Freiberg A 2002. Pathways to justice: sentencing review 2002. Melbourne:
Sentencing Advisory Council
Freiberg A 2003. The four pillars of justice: A review essay. Australia and New
Zealand Journal of Criminology 36: 223-230
Freiberg A 2006. Twenty years of change in the sentencing environment and courts‟
responses. Paper to the Sentencing conference: Principles, perspectives and
possibilities. Canberra: 2006
French V 2007. Juries – a central pillar or an obstacle to a fair and timely criminal
justice system? Reform 90: 40-42
Garland D 2000. The culture of high crime societies: Some preconditions of recent
“Law and Order” politics. British Journal of Criminology 40(3): 347-375
98
References
Garland D 2001. The culture of control. Oxford: Oxford University Press
Gaubatz K 1994. Crime in the public mind. Ann Arbor: University of Michigan Press
Gelb K 2006. Myths and misconceptions: public opinion versus public judgment
about sentencing. Melbourne: Sentencing Advisory Council
Gelb K 2008a. More myths and misconceptions. Melbourne: Sentencing Advisory
Council
Gelb K 2008b. Measuring public opinion about sentencing. Melbourne: Sentencing
Advisory Council
Gleeson AM 2005. Out of touch or out of reach? The Judicial Review 7: 241-253
Gleeson AM 2007. Public confidence in the courts. Paper to Confidence in the courts
conference, National Judicial College of Australia Canberra, 9th February 2007
Graber D 1980. Crime, news and the public. New York: Praeger
Green D 2006. Public opinion versus public judgment about crime: Correcting the
„Comedy of Errors‟. British Journal of Criminology 46(1): 131-154
Horan J & Tait D 2007. Do juries adequately represent the community? A case study
of civil juries in Victoria. Journal of Judicial Administration 16:179-199
Hough M 1996. People talking about punishment. The Howard Journal 35: 191-214
Hough M & Roberts JV 1998. Attitudes toward punishment: Findings from the
British Crime Survey. Home Office Research Study No. 179. London: Home Office
Hough M & Roberts JV 1999. Sentencing trends in Britain: Public knowledge and
public opinion. Punishment and Society 1(1): 11-26
Hough M & Roberts JV 2004. Confidence in Justice: An international review.
London: Institute of Criminal Policy Research
Hough M & Roberts JV 2007. Public opinion, crime and criminal justice: The British
Crime Survey and beyond, in Hough M & Maxfield M (eds), Surveying crime in the
21st Century. Cullompton: Willan Publishing: 197-220
Hutton N 2005. Beyond populist punitiveness? Punishment and Society 7(3): 243
ICM Research 2006. Victims of crime survey. London: Smart Justice and Victim
Support
Indermaur D 1987. Public perception of sentencing in Perth, Western Australia.
Australian and New Zealand Journal of Criminology 20: 163-183
Indermaur D 1990. Crime seriousness and sentencing: A comparison of court practice
and the perceptions of a sample of the public and judges. Canberra: Report to the
Criminology Research Council.
http://www.criminologyresearchcouncil.gov.au/reports/32-89.pdf
99
Jury Sentencing Survey
Indermaur D 2000. Voodoo politics in the era of the TV game show: Public opinion,
the media and political decision making. Presentation to the Centre for Criminology
at the University of Hong Kong and the Hong Kong Criminology Society
Indermaur D & Hough M 2002. Strategies for changing public attitudes to
punishment, in Roberts J & Hough M (eds), Changing attitudes to punishment: Public
opinion, crime and justice. Cullompton: Willan Publishing
Indermaur D & Roberts L 2003. Drug Courts in Australia: What can they offer and
how do we know? Current Issues in Criminal Justice 15(2): 136-154
Indermaur D & Roberts L 2005. Perceptions of crime and justice, in Wilson S,
Meagher G, Gibson R, Denemark D & Western M (eds), Australian social attitudes:
the first report. Sydney: UNSW Press: 141-160
Indermaur D & Roberts L 2009. Confidence in the criminal justice system. Trends
and issues in criminal justice no. 387
Israel M 2000. „What Works‟ with South Australian newspapers. Current Issues in
Criminal Justice 12: 227-232
Jones C, Weatherburn D & McFarlane K 2008. Public confidence in the New South
Wales criminal justice system. Crime and Justice Bulletin: Contemporary Issues in
Crime and Justice 118
Kennamer J 1992. Public opinion, the press and public policy: An introduction, in
Kennamer J (ed), Public opinion, the press and public policy. Westport, CT: Praeger:
1-18
Lasry L 2009. Unelected judges: Out of touch with the community? Blackburn
Lecture delivered to the Law Society of the Australian Capital Territory.
http://www.icjaust.org.au/images/stories/documents/Blackburn_lecture_12.5.09.doc
Lovegrove A 1998. Judicial sentencing policy, criminological expertise and public
opinion. Australian and New Zealand Journal of Criminology 31(3): 287
Lovegrove A 2006. Judges, sentencing and experimental psychology. Journal of
Community Psychology 14(3): 253-266
Lovegrove A 2007. Public opinion, sentencing and lenience: An empirical study
involving judges consulting the community. Criminal Law Review 2007: 769-781
Lushkin R, Fishkin J & Jowell R 2002. Considered opinions: deliberative polling in
Britain. British Journal of Political Science 32: 455-487
Mackenzie G 2005. How judges sentence. Sydney: Federation Press
Maruna S & King A 2004. Public opinion and community penalties, in Bottoms A,
Rex S & Robinson G (eds), Alternatives to prison: Options for an insecure society.
Cullompton: Willan Publishing: 83-111
Mason AM 2002. The courts and public opinion. Bar News March: 30-36
100
References
Mirrlees-Black 2001. Confidence in the criminal justice system: Findings from the
2000 British Crime Survey. Research Findings No. 137. London: Home Office
O‟Brien K, Goodman-Delahunty J, Clough J & Pratley J 2008. Factors affecting juror
satisfaction and confidence in New South Wales, Victoria and South Australia.
Trends and issues in criminal justice no. 354
http://www.aic.gov.au/publications/tandi2/tandi354.html
Page B, Wake R & Ames A 2004. Public confidence in the criminal justice system:
Research Findings No. 221. London: Home Office, Research and Statistics
Directorate
Pratt J 2002. Punishment and civilisation. London: Sage Publications
Roberts JV 1992. Public opinion, crime and criminal justice, in Tonry M (ed), Crime
and justice: A review of research, vol. 16. Chicago: University of Chicago Press
Roberts JV 2007. Public confidence in criminal justice in Canada: A comparative and
contextual analysis. Canadian Journal of Criminology and Criminal Justice April:
153-184
Roberts JV, Crutcher N & Verbrugge P 2007. Public attitudes to sentencing in
Canada: Exploring recent findings. Canadian Journal of Criminology and Criminal
Justice 49: 75-108
Roberts JV & Doob AN 1990. News media influence on public views of sentencing.
Law and Human Behavior 14: 451-468
Roberts JV & Hough M 2005. Understanding public attitudes to criminal justice. UK:
Open University Press
Roberts LD & Indermaur D 2007. Predicting punitive attitudes in Australia.
Psychiatry, Psychology and Law 14(1): 56-65
Roberts JV & Stalans L 1997. Public opinion, crime and criminal justice. Boulder,
CO: Westview Press
Roberts JV, Stalans L, Indermaur D & Hough, M 2003. Penal populism and public
opinion: Lessons from five countries. Oxford: Oxford University Press
Roberts LD & Indermaur D 2009. What Australians think about crime and justice:
Results from the 2007 Survey of Social Attitudes. Australian Institute of Criminology
Research and Public Policy Series no. 101
http://www.aic.gov.au/publications/current%20series/rpp/100-120/rpp101.aspx
Ryan M 2005. Engaging with punitive attitudes towards crime and punishment: Some
strategic lessons from England and Wales, in Pratt J, Brown D, Brown M, Hallsworth
S & Morrison W (eds), The new punitiveness: Trends, theories, perspectives.
Cullompton: Willan Publishing: 139–149
Sackville R 2005. The judiciary and the media. Response to the Australian Press
Council Annual Address, 31 March 2005
101
Jury Sentencing Survey
Sackville R 2009. Let truth and falsehood grapple. Freedom of Speech Conference, 24
March 2009, Sydney
Salisbury H 2004. Public attitudes to the criminal justice system: The impact of
providing information to British Crime Survey Respondents. Home Office On-line
Report 64/04 http://www.homeoffice.gov.uk/rds/pdfs04/rdsolr6404.pdf
Schulz PD 2008. Rougher than usual media treatment: A discourse analysis of media
reporting and justice on trial. Australian Journal of Judicial Administration 17: 223236
Spigelman J 2005. The power of twelve: A new way to sentence for serious crime.
Reform 86: 51-55
Sprott JB & Doob AN 1997. Fear, victimisation and attitudes to sentencing, the courts
and police. Canadian Journal of Criminology 39(3): 275-292
Supreme Court of Tasmania. You be the judge
http://www.courtlists.tas.gov.au/Sentencing/home.html
Thorpe K & Hall P 2009. Chapter 5: Public perceptions in Walker A, Flatley J,
Kershaw C & Moon D (eds) Crime in England and Wales (2008/09). Vol 1: Findings
from the British Crime Survey and police recorded crime. London: HMSO.
Tonry M 1996. Sentencing Matters. New York: Oxford University Press
Tonry M 1999. Rethinking unthinkable punishment policies in America. UCLA Law
Review 46:1751-1791
Tyler TR & Boeckmann RJ 1997. Three strikes and you are out, but why? The
psychology of public support for punishing rule breakers. Law and Society Review 31:
237-264
Victorian Law Reform Commission 1997. Jury service in Victoria, final report vol 3
Walker J, Collins M & Wilson P 1987. How the public sees sentencing: an Australian
survey. Trends and Issues in crime and criminal justice no. 4
Warner CA 2007. Sentencing Review 2006-2007. Criminal Law Journal 31(6): 359368
Wilson P & Brown JW 1973. Crime and the community. Queensland: University of
Queensland Press
Yankelovich D 1991. Coming to public judgement: Making democracy work in a
complex world. Syracuse NY: Syracuse University Press
Young W 2000. An inside look at jury-decision making. Judicial Officers Bulletin
12(4): 25-32
Young W, Cameron N & Tinsley Y 1999. Juries in criminal trials, part two: a
summary of research findings. Wellington: Law Commission of New Zealand
Zander M & Henderson P 1993. The Royal Commission on Criminal Justice: Crown
Court Study. London: HMSO
102
References
Juries Act 2003 (Tas)
Hales v Tasmania [2009] TASSC 100
Tasmania v Allison Dawn Ridley (Unreported, TASSC 8 October 2008)
Tasmania v Cooper (Unreported, TASSC 12 September 2007)
Tasmania v Hume (Unreported, TASSC 5 February 2009)
Tasmania v Ian Anthony Martin (Unreported, TASSC 21 February 2008)
Tasmania v Leslie Charles Rogers; Tasmania v Luke Rodgers (Unreported, TASSC
12 August 2008)
State of Tasmania v Aaron Barry Morley (Unreported, TASSC 22 August 2008)
State of Tasmania v Jaya Krishna Pannala (Unreported, TASSC 12 December 2008)
103
APPENDIX 1
CONSENT FORM
JURY SENTENCING SURVEY
1. I have read and understood the ‘Information Sheet’ for this study.
2. The nature of the study has been explained to me.
3. I understand that this study involves three stages: first, responding to
attached questionnaire (Questionnaire 1); secondly, responding to a
second questionnaire (Questionnaire 2) after I have read some
information including the judge’s sentence and comments. The third stage
involves interviews by a member of the research team of a sample of
those responding to Questionnaire 2. You will have the opportunity to
consent or decline an interview later. There are no foreseeable risks
involved in participating in this study
4. I understand that all research data will be securely stored on the
University of Tasmania premises for 5 years and will then be destroyed.
5. Any questions that I have asked have been answered to my satisfaction.
6. I agree that research data gathered from me may be published provided I
cannot be identified as a participant.
7. I understand that the researchers will keep my identity confidential and
that any information I supply will be used only for the purposes of the
research.
8. I agree to participate in this study by answering the questions on the
attached questionnaire.
9. By writing my postal address and contact details on the questionnaire I
also agree to being sent the judge’s sentencing comments, some
sentencing information and Questionnaire 2.
10. I understand that I may withdraw at any time from this study and that if I
wish, I may request any data I have supplied to date be withdrawn from
the research.
Name of Participant:
Signature:
104
Date:
Appendix 1
Dear Juror,
You are invited to take part in a jury sentencing research project conducted by a
team of researchers from the University of Tasmania. The principal researchers are
Professor Kate Warner, Dr Julia Davis, Dr Maggie Walter and Dr Rebecca Bradfield.
The research has been approved by the University Research Ethics Committee and
the Chief Justice and judges of the Supreme Court of Tasmania. It is being funded
by the Australian Criminology Research Council.
What is the purpose of this research?
It is:



To ascertain public attitudes to sentences imposed by the Courts
To ascertain and improve public knowledge of how the sentencing
system works and of crime trends
To ascertain the level of confidence in the criminal sentencing system
What do I have to do?
Jurors are being asked to take part because they will have heard all the evidence in
the case and thus will be fully informed. The research will involve three separate
stages.



Stage 1 asks you to sign a consent form and to answer some short
questions on the attached form entitled "Jury Sentencing Survey
Questionnaire 1”.
Stage 2 asks you to complete a questionnaire where you can review
your sentence after you have read some sentencing information
including the judge’s sentencing comments. It will also ask you some
related questions about the criminal justice system.
Stage 3: respondents who complete the stage 2 questionnaire will be
asked whether they are willing to take part in a face-to face interview,
further exploring their views as to sentencing issues.
Frequently Asked Questions.

Will the survey be anonymous?
Although we need your name and address in order to send you the
material for Stage 2, the published results will not identify any
participant.

How will you ensure my responses remain confidential?
The questionnaires and interviews will be coded for data entry e.g.
Juror 45. Any material with your name (such as Questionnaire 1 and
the consent form) will remain in a locked cabinet at the University for 5
years and then destroyed.

Do I have to take part in the research?
No. Participation is entirely voluntary. Obviously the more jurors who
take part, the better the results.
105
Jury Sentencing Survey

Will I be disclosing jury deliberations?
No. No question will seek an answer that will involve you disclosing
the deliberations in the jury room.

Who can I contact if I have any questions?
Professor Warner: Ph 62262067; email: kate.warner@utas.edu.au
Dr Davis: email: julia.davis@utas.edu.au
Dr Bradfield: email: rebecca.bradfield@utas.edu.au

Who can I contact if I have any complaints?
Any concerns of complaints about the conduct of the research should
be directed to the Ethics Executive Officer on 03 62267479 or
human.ethics@utas.edu.au

How can I find out the results of this research?
Publication details about the research will be provided on the Law
Faculty’s website at www.law.utas.edu.au
We would very much appreciate your assistance with this project.
Yours sincerely
Professor Kate Warner
(on behalf of the research team)
106
Appendix 1
107
Jury Sentencing Survey
108
Appendix 1
109
Jury Sentencing Survey
110
Appendix 1
111
Jury Sentencing Survey
112
APPENDIX 2
Assault police: sentencing statistics
Assaulting a police officer may be tried in the Magistrates Courts or in the Supreme Court.
More serious charges are heard in the Supreme Court. The table below shows the range of
sentences for one count and global or aggregate sentences when the sentence is imposed for
more than one count. It shows that most sentences are custodial and that the median global
sentence is 5 months but has ranged from 1 month to 2 years. A sentence of 21 months was
imposed on an offender in the 1990-2000 period where an offender with relevant priors
assaulted a hotel licensee and investigating police officers, apparently knocking one
unconscious and punching another a number of times, fracturing his cheek bone and breaking
his nose. About a third of custodial sentences are wholly suspended.
Assaulting police: Custodial Sentences 1978-2008
year
single/global
no
min
med
max
% cust
1978-89
single
12
14d
3m
18m
66
global
11
3m
6m
24m
92
single
2
4m
-
12m
100
global
11
1m
6m
21m
92
single
10
1m
5m
24m
100
global
6
1m
5m
6m
83
1990-00
2001-08
113
Jury Sentencing Survey
Indecent assault (two counts): sentencing statistics
Global sentences for two counts (usually two counts of indecent assault or one of indecent
assault and one of an indecent act with a young person are shown below. The data shows that
in recent years the majority of sentences have been short custodial sentences and that the
median sentence is around 6 months. In the period 2001 to 2008 about 55% of custodial
sentences were wholly suspended. Non-custodial sentences such as fines are not unknown for
this offence.
Indecent assault: Custodial Sentences (Two counts) 1990-2006
114
year
no
min
med
max
% cust
1990-2000
16
2
6
15
94
2001-2008
18
2
5/6
30
86
Appendix 2
UNLAWFULLY INJURING PROPERTY: sentencing statistics
As the table below indicates there are relatively few convictions for injury or
damage to property in the Supreme Court. This is because this offence is usually
heard in the Magistrates Court. However, where the offence is particularly serious
the charge will be laid under the Criminal Code and the matter will heard in the
Supreme Court.
Most sentences in the Supreme Court for this crime are custodial with a median
sentence of 6 months.
Using a car as a weapon to ram into buildings has attracted sentences towards
the top of the range. The sentence of 24 months in the second period was
imposed on an offender, who with co-offenders, stole three vehicles and used
them to in a ‘demolition derby’ causing considerable damage. The fifteen-month
sentence in the 2001-2008 period was imposed on an offender for an attempt to
damage a government building containing forensic evidence which failed but filled
the building with petrol vapour creating a dangerous situation.
Unlawfully Injuring Property: Custodial Sentences (Single count)
year
no
% cust
min
med
max
1978-1989
18
58
1
6
12
1990-2000
17
53
1
6
24
2001-2008
14
57
3
5/6
15
115
Jury Sentencing Survey
STEALING: sentencing statistics
For one count of stealing, statistics show that between 60 and 80 per cent of sentences
imposed have been custodial (this includes suspended prison sentences). The median
sentence is 6 months. Warner (Sentencing in Tasmania, 2002 at p 344) states that in cases of
stealing in the course of employment where between $10,000 and $50,000 is stolen, the
sentences are invariably custodial and range from 4 months to 2 years with a median of 15
months. The longest sentence, one of 6 years 6 months (78 months), was imposed on a
solicitor who pleaded guilty to stealing more than $3million from his clients.
Stealing Custodial Sentences (Single count) 1983-2008
year
no
% cust
min
med
max
1983-1989
74
61
stroc*
6
45
1990-2000
85
77
1
6
32
2001-2008
53
85
1
6
78
*sentenced to the rising of the court (this is technically a sentence of imprisonment that last s until
the court adjourns)
116
Appendix 2
117
Jury Sentencing Survey
118
Appendix 2
119
Jury Sentencing Survey
120
Appendix 2
121
Jury Sentencing Survey
122
Appendix 2
123
Jury Sentencing Survey
124
Appendix 2
125
Jury Sentencing Survey
126
Appendix 2
127
Jury Sentencing Survey
128
APPENDIX 3
129
Jury Sentencing Survey
130
Appendix 3
131
Jury Sentencing Survey
132
Appendix 3
133
Jury Sentencing Survey
134
Appendix 3
135
Jury Sentencing Survey
136
Appendix 3
137
Jury Sentencing Survey
138
Appendix 3
139
Jury Sentencing Survey
140
Appendix 3
141
Jury Sentencing Survey
142
Appendix 3
143
Jury Sentencing Survey
144
Appendix 3
145
Jury Sentencing Survey
146
Appendix 3
147
Jury Sentencing Survey
148